TM

THE REQUISITES
THE REQUISITES
THE REQUISITES
THE REQUISITES
THE REQUISITES
THE REQUISITES is a propieary trademark
of Mosby, Inc.

SERIES EDITOR James H. Thrall, MD

Radiologist-in-Chief
Massachusetts General Hospital
Juan M.Taveras Professor of Radiology
Harvard Medical School Department of Radiology
Boston, Massachusetts

OTHER VOLUMES IN THE REQUISITES Breast Imaging

IN RADIOLOGY SERIES
Cardiac Imaging
Gastrointestinal Imaging
Genitourinary Imaging
Musculoskeletal Imaging
Neuroradiology
Nuclear Medicine
Pediatric Imaging
Ultrasound
Thoracic Imaging
Vascular & Interventional Imaging
1600 John F. Kennedy Boulevard, Suite 1800
Philadelphia, PA 19103-2899

NUCLEAR MEDICINE:THE REQUISITES IN RADIOLOGY ISBN: 978-0-323-02946-9

ISBN: 0-323-02946-9

Copyright 2006, 2001, 1995 Mosby, Inc. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or
by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior permission of
the publisher (WB Saunders, 1600 John F. Kennedy Boulevard, Suite 1800, Philadelphia, PA 19103-2899).

Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia,
PA, USA: phone: (+1) 215 239 3804, fax: (+1) 215 239 3805, e-mail: healthpermissions@elsevier.com.You
may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by
selecting ‘Customer Support’ and then ‘Obtaining Permissions’.

Notice

Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or
appropriate. Readers are advised to check the most current information provided (i) on procedures
featured or (ii) by the manufacturer of each product to be administered, to verify the recommended
dose or formula, the method and duration of administration, and contraindications. It is the responsi-
bility of the practitioner, relying on his or her own experience and knowledge of the patient, to make
diagnoses, to determine dosages and the best treatment for each individual patient, and to take all
appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editor
assumes any liability for any injury and/or damage to persons or property arising out or related to
any use of the material contained in this book.

Third Edition

Library of Congress Cataloging-in-Publication Data

Ziessman, Harvey A.
Nuclear medicine : the requisites / Harvey A. Ziessman, Janis P. O’Malley, James H.Thrall.–3rd ed.
p. ; cm. – (Requisites in radiology)
Thrall’s name appears first on the earlier ed.
Includes bibliographical references and index.
ISBN 0-323-02946-9
1. Nuclear medicine. I. O’Malley, Janis P. II.Thrall, James H. III.Title. IV. Series.
[DNLM: 1. Radionuclide Imaging. 2. Diagnostic Techniques, Radioisotope. 3. Nuclear Medicine–meth
ods.WN 203 Z67n 2006]
R895.T498 2006
616.07’575–dc22 2005050503

Acquisitions Editor: Meghan McAteer

Developmental Editor: Karen Lynn Carter
Publishing Services Manager: Joan Sinclair
Project Manager: Mary Stermel
Marketing Manager: Emily Christie

Printed in the United States

Last digit is the print number: 9 8 7 6 5 4 3 2 1


The 2006 third edition of Nuclear Medicine: The
Requisites in Radiology closely follows the philosophy
and format that has been successful in two prior edi-
tions. Our principal aim has been to provide a concise
introduction and review of the field of nuclear medicine.
Although the textbook has been directed specifically
toward radiology and nuclear medicine residents, it has
found a much wider audience of practicing physicians,
basic scientists, technologists, and medical students.
Dramatic changes have occurred in the field of nuclear
medicine since the publication of the second edition in
2001. New radiotracers have been developed and many
have become clinically important. Radiopharmaceuticals
that depict various aspects of physiology and function
continue to be the cornerstone and unique advantage of
nuclear medicine. Rapid advances in instrumentation
have also powered recent growth in our field.
The use of a radiolabeled glucose analog, F-18 fluo-
rodeoxyglucose (FDG), in conjunction with positron emis-
sion tomography (PET) for tumor imaging,has dramatically
changed the practice of nuclear medicine.The acceptance
and enthusiasm for PET by oncologists, surgeons, and radi-
ation therapists has made FDG PET imaging mandatory for
state-of-the-art oncologic care of many cancers.
Medicare and insurance reimbursement for FDG PET
scanning resulted in an exponential sales growth of PET
scanners in the late 1990s and early 2000s. However, we
have now seen a dramatic shift from dedicated PET cam-
eras to hybrid PET/CT systems that allow acquisition of
both in a single study. In the second edition, we added
a basic science chapter on SPECT and PET because of its
growing importance. In the third edition we have added
a new clinical chapter dedicated to the subject of onco-
logic F-18 FDG PET and PET/CT.
PET is also beginning to show growth in areas other
than oncology, notably, nuclear cardiology, and neurol-
ogy. For example, the use of Rb-82 myocardial perfusion
imaging is in an early growth phase. F-18 FDG has
Preface
become the gold standard for the diagnosis of myocar-
dial viability. Neurologic PET, long important for the
investigation of pathophysiologic mechanisms, is
becoming increasingly important clinically with aware-
ness and concern for the diagnosis and therapy of
dementias.
Although single-photon nuclear cardiology is now
a mature methodology, SPECT myocardial perfusion imag-
ing continues to grow at a very rapid rate, exceeded only
by FDG PET. A complete cardiac study now routinely
includes analysis of myocardial perfusion using SPECT,
gated wall motion and thickening, left ventricular ejec-
tion fraction, attenuation correction, and various two-
and three-dimensional quantitative displays.
Single-photon oncologic nuclear medicine is also an
important area of growth and change. Peptide imaging with
radiolabeled somatostatin receptor imaging agents has
become a routine study in many nuclear medicine practices,
requested for the evaluation and follow-up of patients with
neuroendocrine tumors. Radiolabeled monoclonal anti-
bodies have matured and are used clinically for both
diagnosis and therapy. Radioimmunotherapy is coming
into its own with growing clinical demand for I-131– and
Y-90–labeled monoclonal antibodies for therapy of B-cell
lymphoma. A new infection-seeking antibody radiophar-
maceutical has recently become available that does not
require in vitro labeling, which likely will become a valu-
able diagnostic imaging agent without the potential dan-
gers of blood handling.
Nuclear Medicine: The Requisites has always empha-
sized basic principles and concepts of radiopharmaceu-
ticals and instrumentation. An appreciation of the
pharmacokinetics, distribution, and clearance of radio-
pharmaceuticals and the power and limitations of mod-
ern single- and dual-photon cameras, in conjunction with
an understanding of pathophysiology, provides the tools
necessary for choosing optimal imaging protocols and
for image interpretation.
v
vi PREFACE
Many chapters have been completely rewritten, most
notably, Nuclear Cardiology, Neurology, Endocrine,
and Genitourinary. Others have been significantly revised
and updated (e.g., Skeletal, Single-Photon Oncology,
Pulmonary, Infection and Inflammation, Hepatobiliary,
and Gastro-enterology). The popular Pearls, Pitfalls, and
Frequently Asked Questions has been expanded and
updated.
Important principles continue to be highlighted in
the boxes and tables in each chapter. Typical protocols
are presented for each study type. Schematic diagrams
illustrate basic concepts and innumerable scintigraphic
examples are provided. Many new figures, tables, and
boxes have been added in this edition.
A new co-author,Janis P.O’Malley,MD,has given the text-
book new energy and a fresh perspective.We hope that you
find the third edition of the Nuclear Medicine: The
Requisites to be even better than the two previous editions
and that it is a continual source of knowledge and ideas,
presented is an easy to read and understandable manner.
Harvey A. Ziessman
Janis P. O’Malley
James H. Thrall

We would like to thank those persons that have con-
tributed to the preparation of Nuclear Medicine: The
Requisites, third edition. First and foremost, our gratitude
goes out to our many trainees, who over the years who
have taught us at least as much as we have taught them.
Some have helped in the preparation and review of this
third edition. We thank: William Lavely, MD, Heather A.
Jacene, MD, Alex Dibona, MD, Matt Larrison, MD, Joel
Mixon, MD, Carl Merrow, MD, Robert Morris, MD, Ashok
Muthukrishnan, MD, and Jubal Watts, MD. Many col-
leagues have contributed in direct and indirect ways.We
particularly wish to thank Frederic H. Fahey, D.Sc, Jon
Baldwin, DO, Honggang Liu, MS, Eva V. Dubovsky, MD,
Sharon White, PhD, Douglas F. Eggli, MD, Massoud Madj,
Acknowledgments
MD, James M. Mountz, MD, PhD, and Christopher
J. Palestro, MD.
New high-quality original diagrams and illustrations
have been provided by Tim Phelps and Kate Weaver in
the Johns Hopkins Department of Art as Applied to
Medicine, Rick Tracy in the Department of Pathology at
Johns Hopkins, and David Fisher in the UAB Medical
Education and Design Services.
We would also like to thank our families for support-
ing us in this endeavor and enduring the many hours of
our absences from them to complete this edition. We
have been stimulated and taught by our past mentors,
our present colleagues, and many trainees.Thus we ded-
icate the third edition to all of them.
vii

The third edition of Nuclear Medicine: The Requisites
continues to follow the philosophy and format of the first
two editions.The basic science chapters are designed to
present important principles of physics, instrumentation,
and nuclear pharmacy in the context of how they help
shape clinical practice.The clinical chapters continue to
follow a logical progression from basic principles of
tracer distribution and localization to practical clinical
applications. The authors believe that understanding
tracer mechanisms is fundamental to nuclear medicine
practice and that knowledge of how radiopharmaceuti-
cals localize temporally and spatially is the best deductive
tool available for analyzing images rather than simply
memorizing representative illustrations.
Both the basic science and the clinical chapters have
been extensively revised and rewritten to reflect the chang-
ing nature of nuclear medicine practice.At the time the first
edition of Nuclear Medicine:The Requisites was published,
single photon computed emission computed tomogra-
phy (SPECT) was not in ubiquitous use and positron emis-
sion computed tomography (PET) was restricted to a small
number of research centers.Today,barely a decade later,the
use of SPECT is widespread especially for cardiac applica-
tions and PET has moved well beyond the halls of acade-
mia. SPECT and PET applications now dominate nuclear
medicine practice.These methods are fully described and
richly illustrated throughout the book.
The basic chapter describing SPECT and PET now
also includes material on PET-CT, which is rapidly becom-
ing an important and even dominant new approach to
correlative imaging. PET-CT brings the functional and
metabolic information available from nuclear medicine
studies together with the exquisite anatomical detail
available from computed tomography. New pharmaceu-
ticals have been added to the nuclear medicine arma-
mentarium and are included as appropriate in the organ
system chapters. Keeping up with the manifold changes
occurring in the specialty is challenging to practitioners
Foreword
of nuclear medicine and presented formidable chal-
lenges to the authors of Nuclear Medicine: The
Requisites in making sure that each chapter reflects cur-
rent state-of-the-art practices.
The Requisites in Radiology have now become old
friends to a generation of radiologists.The original intent
of the series was to provide the resident or fellow with
a text that might be reasonably read within several days
at the beginning of each subspecialty rotation and per-
haps reread several times during subsequent rotations or
during board preparation.The series is not intended to be
exhaustive but to provide the basic conceptual, factual,
and interpretive material required for clinical practice.
Each book in The Requisites series is written by
nationally recognized authorities in their respective sub-
specialty areas. Each author is challenged to present
material in the context of today’s practice of radiology
rather than grafting information about new imaging
modalities onto old out-of-date material. It is our hope in
adopting this strategy that readers will find The
Requisites to be a very efficient way of accessing the
most important material.
The first two editions of Nuclear Medicine: The
Requisites were well received in the radiology and
nuclear medicine community, and we are hopeful that
the third edition will be regarded as equally outstanding.
We hope that Nuclear Medicine: The Requisites will
serve residents in radiology as a concise and useful intro-
duction to the subject and will also serve as a very man-
ageable text for review by fellows and practicing nuclear
medicine specialists and radiologists.
James H. Thrall, M.D.
Radiologist-in-Chief
Massachusetts General Hospital
Juan M.Taveras Professor of Radiology
Harvard Medical School
Boston, Massachusetts
ix
 1
CHAPTER Radiopharmaceuticals

Radiopharmaceuticals, Radiochemicals, Radiopharmaceuticals portray the physiology, biochem-

and Radionuclides istry, or pathology of a body system without causing any
Design Characteristics of Radiopharmaceuticals perturbation of function. They are referred to as radio-
Production of Radionuclides tracers because they are given in subpharmacologic
Radionuclide Generators doses that “trace” a particular physiological or pathologic
Molybdenum-99/Technetium-99m Generator Systems process in the body.
Generator Operation and Yield Most radiopharmaceuticals are a combination of
Quality Control a radioactive molecule that permits external detection
Radionuclidic Purity and a biologically active molecule or drug that acts
Chemical Purity as a carrier and determines localization and biodistrib-
Radiochemical Purity ution. For a few radiotracers, the radioactive atoms
Technetium Chemistry and Radiopharmaceutical themselves confer the desired localization properties
Preparation and an attached larger pharmaceutical component
Quality Assurance of Technetium-99m–Labeled is not required, such as I-131 or I-123 sodium iodide,
Radiopharmaceuticals gallium-67 citrate (Ga-67), and thallium-201 chloride
Other Single-Photon Agents (Tl-201).
Radioiodines I-131 and I-123 Table 1-1 summarizes some of the localization mecha-
Indium-111 nisms of radiopharmaceuticals important to clinical
Gallium-67 Citrate practice. Understanding the mechanism or rationale
Thallium-201 for the use of each agent is critical to understanding
Radioactive Inert Gases the normal and pathological findings demonstrated
Radiopharmaceuticals for Positron Emission scintigraphically. Because both naturally occurring and
Tomography synthetic molecules can potentially be radiolabeled,
Dispensing Radiopharmaceuticals great flexibility exists in designing and developing radio-
Normal Procedures pharmaceuticals.
Special Considerations Radiopharmaceuticals for each major clinical applica-
Pregnancy and Lactation tion are considered in detail in chapters on the respec-
Dosage Selection for Pediatric Patients tive organ systems. This chapter presents some of the
Medical Event (Misadministration) general principles of radiopharmaceutical production,
Adverse Reactions to Diagnostic Radiopharmaceuticals radiolabeling, quality assurance, and dispensing.
Radiation Accidents (Spills)
Quality Control in the Nuclear Pharmacy
Sterility and Pyrogen Testing RADIOPHARMACEUTICALS,
Radiopharmaceutical Dose Calibrators RADIOCHEMICALS, AND
Accuracy RADIONUCLIDES
Linearity
Precision or Constancy The term radionuclide refers only to radioactive atoms.
Geometry When a radionuclide is combined with a chemical mole-
Radiation Dosimetry cule to confer desired localization properties, the combi-
nation is referred to as a radiochemical. The term
3
4 NUCLEAR MEDICINE: THE REQUISITES

result in gamma emissions of suitable energy (100–200

Table 1-1 Mechanisms of Radiopharmaceutical keV is ideal for gamma cameras) and sufficient abundance
Localization of emissions for external detection. It should not contain
particulate radiation (e.g.,beta emissions),which increases
Mechanism Applications or examples the patient’s radiation dose without adding diagnostic
information. Beta emissions are suitable for therapeutic
Compartmental Blood pool imaging, direct
localization cystography
radiopharmaceuticals. The effective half-life should only
Passive diffusion Blood-brain barrier break- be long enough for the intended application, usually a few
(concentration down, glomerular hours. The specific activity should be high. Technetium-
dependent) filtration, cisternography 99m (Tc-99m) closely matches these desirable features.
Capillary blockade Perfusion imaging of lungs The pharmaceutical component should be free of any
(physical entrapment)
Physical leakage from a Gastrointestinal bleeding,
toxicity or secondary effects. The radiopharmaceutical
luminal compartment detection of urinary tract should not disassociate in vitro or in vivo, should be
or biliary system leakage readily available or easily compounded, and should have
Metabolism Glucose, fatty acids a reasonable cost. The agent should rapidly and specifi-
Active transport (active Hepatobiliary imaging, renal cally localize according to the intended application.
cellular uptake) tubular function, thyroid
and adrenal imaging
Background clearance should be rapid, leading to good
Chemical bonding and Skeletal imaging target-to-background ratios.
adsorption
Cell sequestration Splenic imaging (heat-
damaged red blood cells), PRODUCTION OF RADIONUCLIDES
white blood cells
Receptor binding and Adrenal medullary imaging,
storage somatostatin receptor Naturally occurring radionuclides have long half-lives
imaging (>1000 years) and are heavy, toxic elements (e.g., ura-
Phagocytosis Reticuloendothelial system nium, actinium, thorium, radium, and radon). They have
imaging no clinical role in diagnostic nuclear medicine. The
Antigen-antibody Tumor imaging
radionuclides most commonly used clinically are artifi-
Multiple
mechanisms
cially produced.
Bombardment of medium atomic-weight nuclides
Perfusion and active Myocardial imaging with low-energy neutrons in nuclear reactors (neutron
transport activation) results in neutron-rich radionuclides that
Active transport and Thyroid uptake and imaging undergo beta-minus decay. Neutron bombardment of
metabolism
Active transport and Hepatobiliary imaging,
enriched uranium-235 (U-235) results in fission products
secretion salivary gland imaging in the middle of the atomic chart (Fig. 1-1). The uncon-
trolled release of radioactive iodines in atomic bomb
explosions and during accidents at nuclear power plants
radiopharmaceutical is reserved for radioactive materi- is a well-known phenomenon that can also be used for
als that have met legal requirements for administration to production purposes under controlled conditions in
patients or subjects. In the United States, this approval a nuclear reactor. Proton bombardment of a wide vari-
must be given by the Food and Drug Administration ety of target nuclides in cyclotrons or other special accel-
(FDA) before these substances can be commercially pro- erators produces proton-rich radionuclides that undergo
duced and used for clinical purposes. positron decay or electron capture.
The term carrier-free implies that a radionuclide is not Radionuclides produced in nuclear reactors include
contaminated by either stable or radioactive nuclides of I-131, Xe-133, Cr-51, and molybdenum-99; those pro-
the same element. The presence of carrier material can duced in cyclotrons or particle accelerators include F-18,
influence biodistribution and efficiency of radiolabeling. I-123, Ga-67,Tl-201, and In-111. Tables 1-2 and 1-3 sum-
The term specific activity refers to the radioactivity per marize the production source and physical characteris-
unit weight (mCi/mg). Carrier-free samples of a radionu- tics of commonly used radionuclides in clinical nuclear
clide have the highest specific activity. medicine practice.

Design Characteristics
of Radiopharmaceuticals RADIONUCLIDE GENERATORS
Certain characteristics are desirable in the design of One of the practical issues faced in nuclear medicine is
radiopharmaceuticals. The radionuclide decay should the desirability of using relatively short-lived agents
 Radiopharmaceuticals 5

Figure 1-1 Periodic table.

Table 1-2 Physical Characteristics of Single-Photon Radionuclides Used in Clinical Nuclear Medicine

Principal mode Principal photon energy

Radionuclide of decay Physical half-life in keV (abundance) Production method

Molybdenum-99 Beta minus 2.8 d 740 (12%), 780 (4%) Reactor

Technetium-99m Isomeric transition 6 hr 140 (89%) Generator
(molybdenum-99)
Iodine-131 Beta minus 8d 364 (81%) Reactor
Iodine-123 Electron capture 13.2 hr 159 (83%) Cyclotron
Gallium-67 Electron capture 78.3 hr 93 (37%), 185 (20%), 300 (17%), 395 (5%) Cyclotron
Thallium-201 Electron capture 73.1 hr 69-83 (Hg x-rays), 135 (2.5%), 167 (10%) Cyclotron
Indium-111 Electron capture 2.8 d 171 (90%), 245 (94%) Cyclotron
Xenon-127 Electron capture 36 d 172 (26%), 203 (7%), 375 (17%) Cyclotron
Xenon-133 Beta minus 5.2 d 81 (37%) Reactor
Cobalt-57 Electron capture 272 d 122 (86%) Cyclotron

Table 1-3 Physical Characteristics of Positron-Emitting Radionuclides Used in Nuclear Medicine

Radionuclide Physical half-life (min) Positron energy (MeV) Range in soft tissue (mm) Production method

Carbon-11 20 0.96 4.1 Cyclotron

Nitrogen-13 10 1.19 5.4 Cyclotron
Oxygen-15 2 1.73 7.3 Cyclotron
Fluorine-18 110 0.635 2.4 Cyclotron
Gallium-68 68 1.9 8.1 Generator (germanium-68)
Rubidium-82 1.3 3.15 15.0 Generator (strontium-82)
6 NUCLEAR MEDICINE: THE REQUISITES

Table 1-4 Radionuclide Generator Systems and Table 1-5 Molybdenum-99 (Mo-99)/Technetium-
Parent and Daughter Half-Lives 99m (Tc-99m) Generator Systems

Parent’s Daughter’s Radionuclides Parent (Mo-99) Daughter (Tc-99m)

Parent half-life Daughter half-life
Half-life 66 hr 6 hr
Molybdenum-99 66 hr Technetium-99m 6 hr Mode of decay Beta minus Isomeric transition
Rubidium-81 4.5 hr Krypton-81m 13 sec Daughter products Tc-99m, Tc-99 Tc-99
Germanium-68 270 d Gallium-68 68 min Principal photon 740 keV, 780 keV 140 keV (89%)
Strontium-82 25 d Rubidium-82 1.3 min energies*
Tin-113 115 d Indium-113m 1.7 hr GENERATOR FUNCTION
Yttrium-87 3.3 d Strontium-87m 2.8 hr
Tellurium-132 3.2 d Iodine-132 2.3 hr Composition of Al2O3
ion exchange
column
Eluant Normal saline
(0.9%)
Time from elution 23 hr
(hours rather than days or weeks) and at the same time to maximum
the need to have radiopharmaceuticals delivered to hos- daughter yield
pitals or clinics from commercial sources. One way
*The decay scheme for Mo-99 is complex, with over 35 gamma rays of different
around this dilemma is the use of radionuclide generator
energies given off. The listed energies are those used in clinical practice for
systems. These systems consist of a longer-lived parent radionuclidic purity checks.
and a shorter-lived daughter. With this combination of
half-lives, the generator can be shipped from a commer-
Maximum activity
cial vendor and the daughter product will still have a use- 1.0 at 23 hr
ful half-life for clinical applications. Although a number Mo-99 decay
of generator systems have been developed over the T1/2  2.8 days
years (Table 1-4), the most important one has been the
Relative activity

0.5
Mo-99/Tc-99m system (Table 1-4).
Ingrowth

Tc-99m
MOLYBDENUM-99/TECHNETIUM-99M T1/2  6 hr
0.2 elution Elution Partial
GENERATOR SYSTEMS elution

The historical production method for Mo-99 was a neu-

0.1
tron activation reaction of Mo-98: 10 20 30 40 50 60 70 80
Mo-98 (n, γ) → Mo-99 Time (hr)
Figure 1-2 Decay curve for molybdenum-99 and ingrowth
This older production method resulted in low specific curves for Tc-99m. Successive elutions, including a partial elution
activity, requiring a large ion exchange column to hold are illustrated. Relative activity is plotted on a logarithmic scale,
both the desired Mo-99 and the carrier Mo-98 left over accounting for the straight line of Mo-99 decay.
from the target material. This process resulted in low
specific concentrations of Tc-99m pertechnetate from The loaded column is placed in a lead container with
generator elution because of the large volume of eluant tubing attached at each end to permit column elution.
needed for complete removal of the Tc-99m activity. Commercial generator systems are autoclaved and the
Mo-99 is now produced by the fission of U-235. elution dynamics are quality controlled before shipment.
The product is often referred to as “fission moly.” The Alternatively, systems may be aseptically assembled from
reaction is: previously sterilized components.
U-235 (n, fission) → Mo-99

After Mo-99 is produced in the fission reaction,it is chemi-
cally purified and passed on to an anion exchange column
composed of alumina (Al2O3) (Table 1-5). The column is
typically adjusted to an acid pH to promote binding. The
positive charge of the alumina binds the molybdate ions
firmly.
Generator Operation and Yield
Figure 1-2 illustrates the relationship between Mo-99
decay and the ingrowth of Tc-99m. Maximum buildup of
Tc-99m activity occurs at 23 hours after elution. This
time point is convenient, especially if sufficient Tc-99m
activity is available to accomplish each day’s work.

Lead
shield
30-ml
evacuated
vial
Lead
shield
Mo-99
bound
to
alumina
anion
exchange
column
Filter
500-ml
reservoir
0.9% saline
Figure 1-3 “Wet”radionuclide generator system.
Otherwise, the generator can be eluted, or “milked,”more
than once a day. Partial elution is also illustrated in Fig. 1-2.
Fifty percent of maximum is reached in approximately
4.5 hours and 75% of maximum is available at 8.5 hours.
Although greatest attention is paid to the rate of Tc-99m
buildup, it should also be remembered that Tc-99m is con-
stantly decaying, with buildup of stable Tc-99 (or “carrier”
Tc-99) in the generator. Generators received after com-
mercial shipment or generators that have not been eluted
for several days have significant carrier Tc-99 in the elu-
ate. Because the carrier Tc-99 chemically behaves in an
identical fashion to Tc-99m, it can compete and adversely
affect radiopharmaceutical labeling efficiency. Many
labeling procedures require the reduction of Tc-99m from
a +7 valence state to a lower valence state. If the eluate
contains sufficient carrier Tc-99, complete reduction may
not occur, with resultant poor labeling and undesired
radiochemical contaminants in the final preparation.
There are two basic types of generator systems with
respect to elution. “Wet”systems are provided with a reser-
voir of normal saline (0.9%) (Fig. 1-3). Elution is accom-
plished by placing a special sterile vacuum vial on the exit
or collection port. The vacuum vial is designed to draw the
appropriate amount of saline across the column. In “dry”
systems, a volume-calibrated saline charge is placed on the
entry port and a vacuum vial is placed on the collection
port (Fig. 1-4). The vacuum draws the saline eluant out of
the original vial, across the column, and into the elution
vial. Elution volumes are typically in the range of 5–20 ml.
Elutions can be performed for add-on or emergency studies
that come up in the course of a day (Fig. 1-2).
Radiopharmaceuticals 7
Lead shield
Saline
30-ml
evacuated
vial
Mo-99
bound
to
alumina
anion
exchange
column
Lead shield
Filter
Figure 1-4 “Dry”radionuclide generator system.
From Figure 1-2,it is obvious that the amount of Tc-99m
activity available from a generator decreases each day as
a result of decay of the Mo-99 parent. In practice, the 2.8-
day half-life of Mo-99 allows generators to be used for
2 weeks, although many larger nuclear medicine opera-
tions require two generator deliveries per week.
Quality Control
Although rigorous quality control is performed prior to
commercial generator shipment, it is important that each
laboratory perform quality control steps each time the
generator is eluted (Table 1-6). These quality control
steps are good medical practice and are necessary to
meet various federal and state regulatory guidelines.
Radionuclidic Purity
The only desired radionuclide in the Mo-99/Tc-99m gen-
erator eluate is Tc-99m. Any other radionuclide in the
sample is considered a radionuclidic impurity and is
undesirable because it will result in additional radiation
exposure to the patient without clinical benefit.
The most common radionuclidic contaminant in the
generator eluate is the parent radionuclide, Mo-99. Tc-99,
the daughter product of the isomeric transition of Tc-99m,
8 NUCLEAR MEDICINE: THE REQUISITES
Table 1-6 Purity Checks: Molybdenum-99 (Mo-99)/
Technetium-99m (Tc-99m) Generator
Systems
Problem Standard
Radionuclidic Excessive Mo-99 <0.15 μCi Mo-99/mCi
purity in eluant Tc-99m at time of
administration
Chemical purity Al2O3 from <10 μg/ml (fission
generator ion generator) (aurin
exchange column tricarboxylic acid
in elution spot test)
Radiochemical Reduced oxidation 95% of Tc-99m
purity states of Tc-99m activity should be
(i.e., +4, +5, or +6 in +7 oxidation
instead of +7) state
is also present but is not considered an impurity or contam-
inant. Although Tc-99 can be a problem from a chemical
standpoint in radiolabeling procedures, it is not a problem
from a radiation or health standpoint and is not tested for as
a radionuclide impurity. The half-life of Tc-99 is 2.1 × 105
years. It decays to ruthenium-99,which is stable.
The amount of parent Mo-99m in the eluate should be
as small as possible because any contamination by a long-
lived radionuclide increases the radiation dose without
providing any benefit to the patient. The Nuclear
Regulatory Commission (NRC) sets limits on the amount
of Mo-99 in the eluate and this must be tested on each
elution. Perhaps the easiest and most widely used
approach is to take advantage of the energetic 740- and
780-keV gamma rays of Mo-99 with dual counting of the
specimen. The generator eluate is placed in a lead con-
tainer carefully designed so that all of the 140-keV pho-
tons of technetium are absorbed but approximately 50%
of the more energetic Mo-99 gamma rays can penetrate.
Adjusting the dose calibrator to the Mo-99 setting pro-
vides an estimate of the number of microcuries of Mo-99
in the sample. The unshielded sample is then measured
on the Tc-99m setting and a ratio of Mo-99 to Tc-99m
activity can be calculated.
The NRC limit is 0.15 μCi of Mo-99 activity per 1 mCi
of Tc-99m activity in the administered dose (Table 1-5).
Because the half-life of Mo-99 is longer than that of Tc-99m,
the ratio actually increases with time, which is rarely a
problem. However, if the initial reading shows near maxi-
mum Mo-99 levels, either the actual dose to be given to
the patient should be restudied before administration or
the buildup factor should be computed mathematically.
From a practical standpoint, the Mo-99 activity may be
taken as unchanged and the Tc-99m decay calculated
(Table 1-7). Breakthrough is rare but unpredictable.
Table 1-7 Physical Decay of Technetium-99m
Time (hr) Fraction remaining
0 1.000
1 0.891
2 0.794
3 0.708
4 0.631
5 0.532
6 0.501
7 0.447
8 0.398
9 0.355
10 0.316
11 0.282
12 0.251
Tc-99m physical half-life = 6.02 hours.
When it does occur, Mo-99 levels can be far higher than
the legal limit.
Chemical Purity
Another routine quality assurance step is to measure
the generator eluate for the presence of the column
packing material,Al2O3. For fission generators, the max-
imum alumina concentration is 10 μg/ml. Aurin tricar-
boxylic acid is used for colorimetric spot testing. The
color reaction for a standard 10 μg/ml sample of alu-
mina is compared with a corresponding sample from
the generator eluate. Acceptable levels are present if
the color is less intense than the color of the standard.
The comparison is made visually and qualitatively. No
attempt is made to measure the alumina concentration
quantitatively. Aluminum levels in excess of this limit
have been shown to interfere with the normal distribu-
tion of certain radiopharmaceuticals, resulting in
increased lung activity with Tc-99m sulfur colloid and
increased liver activity with Tc-99m–methylene diphos-
phonate (Tc-MDP).
Radiochemical Purity
When eluted from the generator,the expected valence state
of Tc-99m is +7, in the chemical form of pertechnetate
–
(TcO4 ). The clinical use of sodium pertechnetate as a radio-
pharmaceutical and the preparation of Tc-99m-labeled
pharmaceuticals from commercial kits are based on the +7
oxidation state. The U.S. Pharmacopeia (USP) standard for
the generator eluate is that 95% or more of Tc-99m activity
be in this +7 state. Reduction states at +4,+5,or +6 result in
impurities. However, these reduction states can be
detected by thin-layer chromatography. Problems with
radiochemical purity of the generator eluate are infre-
quently encountered but should be considered if
 Radiopharmaceuticals 9

The details of individual technetium radiopharmaceu-

Table 1-8 Measures of Pharmaceutical Purity ticals are discussed in the chapters on individual organ
systems and include key points in preparation and the
Parameter Definition Example issues recognition of in vivo markers of radiopharmaceutical
impurities. Table 1-9 summarizes the major Tc-99m-
Chemical purity Fraction of wanted Amount of alumina
vs. unwanted breakthrough in
labeled agents that are used clinically.
chemical in Mo-99/Tc-99m Commercial kits contain a reaction vial with the appro-
preparation generator eluate priate amount of stannous ion (tin), the nonradioactive
Radiochemical Fraction of total Amount of bound vs. pharmaceutical to be labeled, and other buffering and
purity radioactivity in unbound Tc-99m in stabilizing agents. The vials are flushed with nitrogen to
desired chemical Tc-99m diphosphonate
form
prevent atmospheric oxygen from interrupting the reac-
Radionuclidic Fraction of total Ratio of Tc-99m vs. tion. Figure 1-5 illustrates the sequence of steps in a sam-
purity radioactivity in Mo-99 in generator ple labeling process. Sodium pertechnetate is drawn into
the form of eluate; I-124 in an a syringe and assayed in the dose calibrator. After the
desired I-123 preparation proper Tc-99m activity is confirmed, the sample is added
radionuclide
Physical purity Fraction of total Correct particle size
to the reaction vial. The amount of Tc-99m activity added
pharmaceutical distribution in for each respective product is determined by the number
in desired Tc-99m MAA of patient doses desired in the case of a multidose vial, an
physical form preparation; absence estimate of the decrease in radioactivity caused by decay
of particulate between the time of preparation and the estimated time
contaminates in any
agent that is a true
of dosage administration, and the in vitro stability of the
solution product. The completed product is labeled and kept in
Biological purity Absence of Sterile, pyrogen-free a special lead-shielded container until it is time to with-
microorganisms preparations draw a sample for administration to a patient. Each patient
and pyrogens dose is individually assayed before being dispensed.
Excessive oxygen can react directly with the stannous
ion, leaving too little reducing power in the kit, which
kit labeling is poor. Measures of pharmaceutical purity are can result in unwanted free Tc-99m pertechnetate in the
summarized in Table 1-8. preparation. A less common problem is radiolysis after
kit preparation, also resulting in free pertechnetate. The
phenomenon is seen when high amounts of Tc-99m
TECHNETIUM CHEMISTRY activity are used. The kit preparations are usually
AND RADIOPHARMACEUTICAL designed so that multiple doses can be prepared from
PREPARATION one reaction vial.

Tc-99m is the most commonly used radionuclide because

of its ready availability, the favorable energy of its princi-
pal gamma photon (140 keV), its favorable dosimetry
with lack of primary particulate radiations, and its nearly
ideal half-life (6 hours) for many clinical imaging studies.
However, technetium chemistry is challenging. In most
labeling procedures, technetium must be reduced from
the +7 valence state. The reduction is usually accom-
plished with stannous ion. One exception is the labeling
of Tc-99m sulfur colloid, which requires heating.
The actual final oxidation state of technetium in many
radiopharmaceuticals is either unknown or subject to
debate. A number of technetium compounds are chelates,
which involve a complex bond at two or more sites on the
ligand. Others are used on the basis of their empirical effi-
cacy without complete knowledge of how technetium is
being complexed in the final molecule. One exception to
the need to reduce technetium from the +7 oxidation state
is in the preparation of Tc-99m sulfur colloid (Tc2S7).
QUALITY ASSURANCE OF
TECHNETIUM-99M–LABELED
RADIOPHARMACEUTICALS
The difficult nature of technetium chemistry highlights
the importance of checking the final product for radio-
chemical purity. This term is defined as the percentage
of the total radioactivity in a specimen that is in the
specified or desired radiochemical form (Table 1-8). For
example, if 5% of the Tc-99m activity remains as free
pertechnetate in a radiolabeling procedure, the radio-
chemical purity would be stated as 95%, assuming no
other impurities. Each radiopharmaceutical has a spe-
cific radiochemical purity to meet USP standards or FDA
requirements, typically 90%. Causes of radiochemical
impurities include poor initial labeling, radiolysis, decom-
position, pH changes, light exposure, or presence of oxi-
dizing or reducing agents.
10 NUCLEAR MEDICINE: THE REQUISITES

Table 1-9 Technetium-99m Radiopharmaceuticals

Agent Application

Tc-99m sodium pertechnetate Meckel’s diverticulum detection, salivary and thyroid gland scintigraphy
Tc-99m sulfur colloid (filtered) Lymphoscintigraphy
Tc-99m sulfur colloid Liver/spleen scintigraphy, bone marrow scintigraphy
Tc-99m pyrophosphate Acute myocardial infarction detection
Tc-99m diphosphonate Skeletal scintigraphy
Tc-99m macroaggregated albumin (MAA) Pulmonary perfusion scintigraphy, liver intra-arterial perfusion scintigraphy
Tc-99m red blood cells Radionuclide ventriculography, gastrointestinal bleeding, hepatic hemangioma
Tc-99m diethylenetriamine-pentaacetic acid (DTPA) Renal scintigraphy, lung ventilation (aerosol), glomerular filtration rate
Tc-99m mercaptoacetyltriglycine (MAG3) Renal dynamic scintigraphy
Tc-99m dimercaptosuccinic acid (DMSA) Renal cortical scintigraphy
Tc-99m iminodiacetic acid (HIDA) derivatives Hepatobiliary scintigraphy
Tc-99m sestamibi (Cardiolite, Miraluma) Myocardial perfusion scintigraphy, breast imaging
Tc-99m tetrofosmin (Myoview) Myocardial perfusion scintigraphy
Tc-99m teboroxime (CardioTec) Myocardial perfusion scintigraphy
Tc-99m exametazime (HMPAO) Cerebral perfusion scintigraphy, white blood cell labeling
Tc-99m bicisate (ECD) Cerebral perfusion scintigraphy
Tc-99m arcitumomab (CEA) Monoclonal antibody for colorectal cancer evaluation
Tc-99m apcitide (AcuTect) Acute venous thrombosis imaging
Tc-99m depreotide (NeoTect) Tumor imaging
Tc-99m fanolesomab (NeutroSpec) Infection imaging

Tc-99m as sodium pertechnetate

from generator eluate

Tc-99m radiopharmaceutical
ready for dispensing
Nitrogen-purged
kit reaction vial
with nonradioactive
materials

A B
Figure 1-5 Preparation of a technetium-99m-labeled radiopharmaceutical. A, Tc-99m as sodium
pertechnetate is added to the reaction vial. B, Tc-99m radiopharmaceutical is ready for dispensing.

In vivo, radiochemical impurities contribute to back-
ground activity or other unwanted localization and
degrade image quality. For many agents, the presence of
a radiochemical impurity can be recognized by altered
in vivo biodistribution. However,intercepting the offend-
ing preparation before administration to a patient is
desirable. A number of systems have been developed to
assay radiochemical purity. The basic approach is to use
thin-layer chromatography. Radiochromatography is per-
formed in the same manner as conventional chromatog-
raphy,by spotting a sample of the test material at one end
of a strip. A solvent is then selected for which the desired
radiochemical and the potential contaminants have
known migration patterns, so the strip can be placed in
 Radiopharmaceuticals 11

Syringe with radiopharmaceutical

for the patient

Plastic shield to
prevent contamination

Patient dose
withdrawn from
vial

Patient dose assayed

Dose
in dose calibrator before
calibrator
it is dispensed

15 mCi

C D
Figure 1-5, cont’d C, The patient dose is withdrawn from the vial. D, Each dose is measured in
the dose calibrator before it is dispensed.

a dose calibrator for counting. The presence of the radio-

label provides an easy means of quantitatively measuring
the migration patterns.
For technetium radiopharmaceuticals, the presence of
free pertechnetate and insoluble, hydrolyzed reduced
technetium moieties are tested using instant thin-layer
chromatography techniques. For example, using acetone
as the solvent, free pertechnetate migrates with the sol-
vent front in a paper and thin-layer chromatography sys-
tem, whereas Tc-99m diphosphonate and hydrolyzed
reduced technetium remain at the origin (Fig. 1-6). For
selective testing of hydrolyzed reduced technetium, a sil-
ica gel strip is used with saline as the solvent. In this sys-
tem, both free pertechnetate and Tc-99m diphosphonate
move with the solvent front and hydrolyzed reduced
technetium again stays at the origin (Fig. 1-6). This com-
bination of procedures allows measurement of each of
Figure 1-6 Radiochromatography for quality control of Tc-99m
the three components. Chromatography systems have diphosphonate. The count-activities on the strips are indicated by
been worked out for each major technetium-labeled the numbers beside the strip diagram. Black dot at the bottom of
radiopharmaceutical. each strip represents the origin. Acceptable (left): 3% of the
Elaborate systems are available to “read” the chro- activity does not migrate with Tc-99m diphosphonate in saline and
matography strips. Chromatographic scanners provide 2% migrates as Tc-99m pertechnetate with the solvent front using
methanol:acetone. Thus, 5% of the radioactivity is not present at
detailed strip chart recording of radioactivity distribu- Tc-99m diphosphonate. Unacceptable (right), 5% of the activity is
tion. In practice, the easiest way to perform chromatog- present as impurities (saline chromatogram) and 22% as free
raphy is simply to cut the chromatography strip into two pertechnetate (methanol:acetone chromatogram). This
pieces that can be counted separately. radiopharmaceutical is of unacceptable quality and should not be
Chromatography is something of an art form and a num- used clinically. The Rf of a compound is the distance from the
center of its activity on the strip to the origin (site of application)
ber of common pitfalls must be avoided. Inadvertently divided by the distance from the solvent front to the origin. An Rf of
immersing the chromatography strip into the solvent past 1.0 means that the compound moves with the solvent front,
the location of the sample spot results in less migration whereas an Rf of 0 means that the component remains at the origin.
12 NUCLEAR MEDICINE: THE REQUISITES

than expected. Also, if spots are not allowed to dry before
being used with organic solvents, spurious migration pat-
terns will occur. On the other hand, excessive delay
before starting the chromatogram can result in reoxida-
tion of the technetium in the sample; again, spurious
results will be encountered.
OTHER SINGLE-PHOTON AGENTS
Radioiodines I-131 and I-123
Radioiodine-131 as sodium iodide was the first radiophar-
maceutical of importance in clinical nuclear medicine. It
was used for routine studies of the thyroid gland for sev-
eral years in the late 1940s (Table 1-10). Subsequently
I-131 was used as the radiolabel for a variety of radiophar-
maceuticals, including human serum albumin, macroag-
gregated albumin, hippuran for renal studies, and adrenal
scintigraphy (metaiodobenzylguanidine and iodocholes-
terol derivatives).
The disadvantages of I-131 include relatively high
principal photon energy (364 keV),long half-life (8 days),
and the presence of beta particle emissions. Radioiodine-
131 remains an important radiopharmaceutical in nuclear
medicine practice for the treatment of hyperthyroidism
and differentiated thyroid cancer.
The quality control of radioiodinated pharmaceuticals
is important to reduce unwanted radiation exposure to
the thyroid gland. In nonthyroid imaging applications of
I-131 as a radiolabel, it is common practice to block the
thyroid gland with oral iodine (e.g., usually supersatu-
rated potassium iodine; occasionally, potassium perchlo-
rate) to prevent thyroid accumulation of any iodine ion
present as a radiochemical impurity or metabolite.
Whenever possible, I-123 is substituted for I-131 for
diagnostic purposes. It has a shorter half-life (Table 1-2)
and its principal photon energy of 159 keV is better
suited to imaging with the gamma scintillation camera.
I-123 decays by electron capture, and the dosimetry is
favorable compared with that of I-131. Even in appli-
cations where imaging over a period of several days
allows for improved target-to-background ratio and thus
I-131would seem advantageous, I-123 is now increasingly
replacing I-131 (e.g.,for whole body thyroid cancer scans
and meta-iodo-benzyl-guanidine [MIBG] imaging for
neuroblastoma and pheochromocytoma).
Indium-111
Another versatile label that has found applications in
clinical nuclear medicine is indium-111 (Table 1-10).
Its principal photon energies of 172 keV and 245 keV

Table 1-10 Radiopharmaceuticals for Single-Photon Imaging (Non-Tc-99m labeled agents)

Agent Application

Tl-201 thallium chloride Myocardial perfusion scintigraphy

Ga-67 gallium citrate
Xe-133 xenon (inert gas)
Xe-127 xenon (inert gas)
Kr-81m krypton (inert gas)
I-131 sodium iodide
I-123 sodium iodide
I-131 hippuran
I-123 hippuran
In-111 oxine leukocytes
I-131, I-123 metaiodobenzyl
guanidine (MIBG)
I-131 NP-59 (6β iodomethyl-19
norcholesterol)
In-111 pentetreotide (OctreoScan)
In-111 capromab pendetide
(ProstaScint)
Sm-153 lexidronam (Quadramet)
Sr-89 chloride
In-111 satumomab pentetretide
(OncoScint)
I-131 tositumomab (Bexxar)
In-111 ibritumomab (Zevalin)
Inflammatory disease detection, tumor imaging
Pulmonary ventilation scintigraphy
Pulmonary ventilation scintigraphy
Pulmonary ventilation scintigraphy
Thyroid cancer scintigraphy; thyroid uptake function studies;
treatment of Graves’ disease, toxic nodule, and thyroid cancer
Thyroid scintigraphy, thyroid uptake function studies
Renal imaging and function studies
Renal imaging and function studies
Inflammatory disease/infection detection
Adrenal medullary imaging, neural crest tumor detection
Adrenal cortical scintigraphy
Somatostatin receptor tumor imaging
Monoclonal antibody for prostate Ca imaging
Bone pain palliation
Bone pain palliation
Monoclonal antibody for colorectal cancer
B-cell lymphoma imaging and therapy
B-cell lymphoma scintigraphy prior to therapy

are favorable compared with I-131. The 2.8-day half-life of
In-111 permits multiple-day sequential imaging. Several
In-111 radiopharmaceuticals have proven clinically use-
ful (e.g., In-111 oxine leukocytes for the evaluation of
inflammatory or infectious disease). The somatostatin
receptor binding peptide, pentetreotide (OctreoScan)
labeled to In-111, binds to a variety of neuroendocrine
tumors. In-111 capromab pendetide (ProstaScint) is
a monoclonal antibody used for detection of recurrent
prostate cancer.
Gallium-67 Citrate
Ga-67 citrate is transported and extracted like iron, local-
izing in tumors and inflammatory conditions. However,
in many respects Ga-67 does not have favorable proper-
ties for scintigraphy. It has multiple photopeaks and the
most abundant photon has the lowest energy (Table 1-2).
In current practice, the lower three photopeaks (93 keV,
185 keV, 300 keV) are acquired. Nonetheless,“downscat-
ter” from the higher energies degrades the image data in
the lower windows.
Other disadvantages of Ga-67 include slow clearance
from background tissues, necessitating delayed imaging
at 48 hours (even ≥72 hours in some applications). Early
excretion (<24 hours) through the kidneys and delayed
excretion via the bowel make imaging in the abdomen
problematic. Care must be taken to interpret the scinti-
graphic images with a full knowledge of how long after
tracer administration the study was obtained. Laxatives
may be required to clear confusing or obscuring activity
from the colon.
Thallium-201
Thallium-201 became clinically available in the mid-
1970s as a radiopharmaceutical for myocardial scintigra-
phy. Thallium behaves as a potassium analog, with high
net clearance (~85%) in its passage through the myocar-
dial capillary bed, which makes it an excellent marker of
regional blood flow to viable myocardium.
The major disadvantage of thallium as a radioactive
imaging agent is the absence of an ideal photopeak for
imaging. The gamma rays at 135 keV and 167 keV
occur in low abundance (Table 1-2). The emitted mer-
cury characteristic x-rays in the range of 69–83 keV
are acquired. The ability of the gamma scintillation
camera to discriminate scattered events from primary
photons is suboptimal at this energy. Because of the
poor imaging characteristics of Tl-201, alternative
Tc-99m labeled radiopharmaceuticals have been
sought and are now available, although they also have
their limitations, as discussed in the chapter on cardiac
imaging.
Radiopharmaceuticals 13
Radioactive Inert Gases
Radioactive inert gases are used for pulmonary ventila-
tion imaging. Xenon-133 (Xe-133) is a convenient agent
to have on hand because of its 5.2-day half-life. Its major
disadvantage is the relatively low energy of its principal
photon (81 keV). This low energy dictates the perform-
ance of ventilation scintigraphy before Tc-99m perfusion
scintigraphy. Nonetheless, Xe-133 is still a commonly
used agent because of its superior distribution in the
lung periphery of patients with chronic obstructive lung
disease.
Xe-127 is theoretically superior to Xe-133 because of
its higher photon energies (Table 1-2). Because its pho-
ton energies are higher than those of Tc-99m, the venti-
lation portion of a ventilation-perfusion study can be
performed after the locations of any perfusion defects
are known, which allows the examination to be tailored
to the findings in individual patients. The high cost of
producing Xe-127 has kept it from wide use.
Krypton-81m has potential advantages because of
its high principal gamma emission of 190 keV and
short half-life of 13 seconds allowing for postperfusion
imaging and multiple view acquisition without con-
cern for retained activity or radiation dose. However,
the rubidium-81/krypton-81m generator system is rela-
tively expensive and not commonly used because it
must be replaced daily due to the generator’s short
half-life.
A host of other radionuclides have been used over
the years. The radionuclides summarized in Fig. 1-1
are the most important in current practice.
RADIOPHARMACEUTICALS FOR
POSITRON EMISSION TOMOGRAPHY
The physical characteristics of commonly used positron-
emitting radionuclides are summarized in Table 1-3.
Many radiopharmaceuticals have been described for use
in positron emission tomography (PET) (Box 1-1).
Carbon, nitrogen, and oxygen are found ubiquitously in
biological molecules. It is thus theoretically possible to
radiolabel just about any molecule of biological interest.
Fluorine-18 (F-18) has the advantage of a longer half-life
than C-11, N-13, or O-15 and has been used as a label for
the glucose analog fluorodeoxyglucose (FDG). F-18 FDG
has found widespread clinical application in whole body
tumor imaging and, to a lesser extent, imaging of the
brain and heart. Tumors derive their energy from glu-
cose metabolism and the uptake of F-18 FDG is a marker
of tumor metabolism and viability.
Rubidium-82 is available from a generator system
with a relatively long-lived parent (strontium-82,T1/2 =
14 NUCLEAR MEDICINE: THE REQUISITES

25 days) (see Table 1-4). Rubidium, like thallium, is

a potassium analog and is used for myocardial perfusion
imaging. Its availability from a generator system obviates
the need for an on-site cyclotron for production. One
limitation of Rb-82 is the high energy (3.15 MeV) of its
positron emissions. This high energy results in a rela-
tively long average path in soft tissue before annihila-
tion, degrading the ultimate spatial resolution available
with the agent. This feature is shared to a lesser extent
by O-15.
The production of most positron-emitting radionu-
clides and their subsequent incorporation into PET radio-
pharmaceuticals is expensive, complex, and requires
a cyclotron or other special accelerator and relatively
elaborate radiochemical-handling equipment. In-house
self-contained small cyclotrons with automated chem-
istry are available but are expensive for most clinical set-
tings. However, the large clinical demand and the
relatively long 2-hour half-life of F-18 FDG has resulted
in its production and distribution by regional radio-
pharmacies.
Box 1-1 Positron Emission Tomography
Selected Radiopharmaceuticals
DISPENSING
RADIOPHARMACEUTICALS
Normal Procedures
General radiation safety procedures should be followed
in any laboratory (Box 1-2). The dispensing of radiophar-
maceuticals is under a series of exacting rules and regula-
tions promulgated by the FDA and NRC, as well as state
pharmacy boards and hospital radiation safety commit-
tees. In brief, radiopharmaceuticals are prescription
drugs that cannot be legally administered without being
ordered by an authorized individual. The nuclear medi-
cine physician and the radiopharmacy are responsible
for confirming the appropriateness of the request, ensur-
ing that the correct radiopharmaceutical in the requested
or designated amount is administered to the patient, and
keeping records of both the request and the documenta-
tion of the dosage administration.
Before any material is dispensed, all appropriate qual-
ity assurance measures should be carried out. These
measures are described in detail earlier in the chapter for
the Mo-99/Tc-99m generator system and Tc-99m-labeled
radiopharmaceuticals. For other agents, the package

PERFUSION AGENTS Box 1-2 Radiation Safety Procedures

Oxygen-15 carbon dioxide
Oxygen-15 water Wear laboratory coats in areas where radioactive
Nitrogen-13 ammonia materials are present.
Rubidium-82 chloride Wear disposable gloves when handling radioactive
materials.
BLOOD VOLUME Monitor hands and body for radioactive contamination
Oxygen-15 carbon monoxide before leaving the area.
Carbon-11 carbon monoxide Use syringe and vial shields as necessary.
Gallium-68 EDTA Do not eat, drink, smoke, apply cosmetics, or store food
in any area where radioactive material is stored or
METABOLIC AGENTS used.
Fluorine-18 sodium fluoride Wear personnel monitoring devices in areas with
Fluorine-18 fluorodeoxyglucose radioactive materials.
Oxygen-15 oxygen Never pipette by mouth.
Carbon-11 acetate Dispose of radioactive waste in designated, labeled, and
Carbon-11 palmitate properly shielded receptacles located in a secured
Nitrogen-13 glutamate area.
Label containers, vials, syringes containing radioactive
TUMOR AGENTS materials.When not in use, place in shielded
Fluorine-18 fluorodeoxyglucose containers or behind lead shielding in a secured area.
Carbon-11 methionine Store all sealed sources (floods, dose calibrator sources)
Fluorine-18 fluorothymidine in shielded containers in a secured area.
Before administering doses to patients, determine and
RECEPTOR-BINDING AGENTS record activity.
Carbon-11 carfentanil Know what steps to take and who to contact (radiation
Carbon-11 raclopride safety officer) in the event of radiation accident,
Fluorine-18 fluoro-L-dopa improper operation of radiation safety equipment, or
Fluorine-18 spiperone theft/loss of licensed material.
 Radiopharmaceuticals 15

insert or protocol for formulation and dispensing should

be consulted to see what radiochromatography or other Table 1-11 Recommendations for Radio-
quality control steps must be performed before dosage pharmaceuticals Excreted in Breast Milk
administration. As a good standard of practice, quality
control should always be performed, even when not Administered
Radiopharma- activity MBq
legally required. Every dose should be physically ceutical (mCi) Counseling Advised
inspected before administration for any particulate or
foreign material, such as bits of rubber from the tops of Ga-67 citrate 185 (5.0) Yes Cessation
multidose injection vials. Each dose administered to I-131 sodium iodide 0.74 (.02) Yes Cessation
a patient must be assayed in a dose calibrator. The admin- I-123 sodium iodide 14.8 (.4) Yes 2-3 d
I-123 MIBG 370 (10) Yes 48 hr
istered activity should be within ±20% of the prescrip- Tl-201 111 (3) Yes 96 hr
tion request. In-111 leukocytes 185 (5) Yes 48 hr
Tc-99m MAA 148 (4) Yes 12 hr
Tc-99m red blood 740 (20) Yes 12 hr
Special Considerations cells in vivo
Tc-99m 185 (5) Yes 4 hr
Pregnancy and Lactation pertechnetate
The possibility of pregnancy should be considered for
every woman of childbearing age referred to the Modified with permission from Stabin MG, Breitz HB: Breast milk excretion of
nuclear medicine service for a diagnostic or therapeutic radiopharmaceuticals: mechanisms, findings, and radiation dosimetry. J Nucl Med
41:863-873, 2000.
procedure. Pregnancy alone is not an absolute con-
traindication to performing a nuclear medicine study.
For example, pulmonary embolism is encountered
in pregnant women and is associated with potential body weight. Empirically, body surface area correlates
serious morbidity and mortality. Thus, the risk-to-bene- better than body weight for dosage selection. Various
fit ratio of ventilation-perfusion scintigraphy is high formulas and nomograms have been developed. Each
and considered a safe procedure in this circumstance. laboratory should select a method and standardize its
The radiation dosage is kept at a minimum. Neither application.
of the radiopharmaceuticals employed (Xe-133 or An approximation based on body weight uses the for-
Tc-99m macroaggregated albumin) crosses the placenta mula:
in considerable amounts. On the other hand, radioio- Patient weight (kg)
dine does cross the placenta. The fetal thyroid develops Pediatric dose = × Adult dose
70 kg
the capacity to concentrate radioiodine at approxi-
mately weeks 10–12 of gestation and cases of cretinism Another alternative is the use of Webster’s rule:
caused by in utero exposure to radioiodine-I-131 have Age + 1
been documented. Pediatric dose = × Adult dose
Age + 7
The management of women who are lactating and
breastfeeding an infant is another special problem. The This formula is not useful for infants. Moreover, in some
need to suspend breastfeeding is determined by the half- cases a calculated dose may not be adequate to obtain
life of the radionuclide involved and the degree to which a diagnostically useful study and physician judgment
it is secreted in breast milk. Radioiodine is secreted by must be used. For example, a newborn infant with sus-
the breast and breastfeeding should be terminated alto- pected biliary atresia may require 24-hour delayed
gether after the administration of I-131. For I-123, it has Tc-99m hepatobiliary iminodiacetic acid (HIDA) imag-
generally been recommended that breastfeeding could ing, which is not feasible if the dose is too low.
safely be resumed after 2–3 days. For Tc-99m agents, Therefore, a minimum dose for each radiopharmaceuti-
12–24 hours is sufficient. Table 1-11 lists recommenda- cal should be established.
tions for the length of time breastfeeding should be
discontinued after administration of various radiophar- Medical Event (Misadministration)
maceuticals. The definition and procedures for handling misadminis-
trations of radiopharmaceuticals are set out in the Code
Dosage Selection for Pediatric Patients of Federal Regulations (10 CFR-35). However, the termi-
A number of approaches have been proposed for scal- nology has changed. What was previously called a mis-
ing down the amount of radioactivity administered to administration is now called a medical event. The code
children. There is no perfect way to do this because of was revised in 2002. Many of the prior “misadministra-
the differential rate of maturation of body organs and tions” no longer have to be reported to the NRC or
the changing ratio of different body compartments to state. A medical event is defined by the NRC rules and
16 NUCLEAR MEDICINE: THE REQUISITES
regulations as a radiopharmaceutical dose admini-
stration involving the wrong patient, wrong radiophar-
maceutical, wrong route of administration, or an
administered dose differing from the prescribed dose
when: (1) the effective dose equivalent to the patient
exceeds 5 rem to the whole body or 50 rem to any indi-
vidual organ (Box 1-3) or (2) a diagnostic dose of I-131
sodium iodide exceeds 30 μCi.
The criteria just described show that medical events
are extremely unlikely to occur as a result of any diag-
nostic nuclear medicine procedure. Most will be
related to radioiodine I-131. However, after the occur-
rence of a medical event is recognized, regulations for
reporting of the event and management of the patient
should be followed. The details are determined in part
by the kind of material involved and the amount of the
adverse exposure of the patient. All medical events
must be reported to the regulatory agency, the referring
physician, and the affected patient. Complete records
on each event must be retained and available for NRC
review for 10 years.
Certain states, called Agreement States, have entered
into regulatory agreements with the NRC that give
them the authority to license and inspect by-product,
source, or special nuclear material used or possessed
within their borders. There are now 32 Agreement
States and this number is growing. Although the NRC
regulates byproduct (reactor-produced) material such
as I-131 and Xenon-133, the states often regulate other
radiation sources as well, such as cyclotron-produced
radioactive material (F-18, Co-57, Ga-67, In-111, and Tl-
201). The states, not the NRC, regulate the use of
positron emission tomography.
Box 1-3 Annual Dose Limits for
Radiation Exposure (NRC
Regulations)
ADULT OCCUPATIONAL
5 rem (0.05 Sv) total effective dose equivalent
50 rem (0.5 Sv) to any organ or tissue or extremity
15 rems (0.15 Sv) to the lens of the eye
MINORS (<18 YEARS OF AGE) OCCUPATIONAL
10% of those for adult workers
EMBRYO/FETUS OCCUPATIONAL
0.5 rem (5mSv) during pregnancy
MEMBERS OF THE PUBLIC
0.1 rem (1 mSv)
2 mrem (0.02 mSv) in any hour (average)
Adverse Reactions to Diagnostic
Radiopharmaceuticals
Adverse reactions to radiopharmaceuticals are much less
common than adverse reactions to iodinated contrast
media. Reactions are usually mild and, for the radiophar-
maceuticals in use today, rarely fatal. The greatest con-
cern of allergic reactions is for agents containing human
serum albumin. Also, preparations of Tc-99m-sulfur col-
loid have a gelatin stabilizer derived from animal protein.
These agents can be associated with allergic reactions.
Of more recent concern is the possibility of reactions
caused by the development of human antimouse anti-
bodies (HAMA) after repeated exposure to radiolabeled
antibody imaging agents. The concern over the develop-
ment of HAMA and potential adverse consequences has
been a factor in the FDA’s slow approval for radiolabeled
antibodies. However, several diagnostic and therapeutic
radiolabeled antibodies have been approved, such as
In-111 ProstaScint,Tc-99m NeutroSpec, In-111 and Y-90
Zevalin.
RADIATION ACCIDENTS (SPILLS)
In a busy nuclear medicine practice handling several
dozen patient doses a day, as well as stock solutions of
generator eluate, with most materials in liquid form, acci-
dental spills of radioactive material occur from time to
time. The spills are somewhat arbitrarily divided into
minor and major categories, depending on the radionu-
clide and the amount spilled. For I-131, incidents involv-
ing activities up to 1 mCi are considered minor; spills
above that level are considered major. For Tc-99m,Tl-201,
and Ga-67, the threshold for considering a spill to be
major is 100 mCi.
The basic principles of responding to both kinds of
spills are the same (Box 1-4). For minor spills, people in
the area are warned that the spill has occurred.
Attempts are made to prevent the spread of the spilled
material. Absorbent paper may be used to cover the
spilled material if it is visibly identifiable. Minor spills
can be cleaned up directly with appropriate technique,
including use of soap and water, disposable gloves, and
remote handling devices. All contaminated material,
including gloves and other objects, should be disposed
of in designated bags. The area should be continually
surveyed until the reading from a Geiger-Müller survey
meter is at background levels. All personnel involved
should also be monitored, including their hands, shoes,
and clothing. The spill should be reported to the insti-
tution’s radiation safety officer.
For major spills, the area is cleared immediately.
Attempts are made to prevent further spread with
absorbent pads, and if possible the radioactivity is
shielded. The room is sealed off and the radiation safety

Box 1-4 Procedure for Radioactive Spill
1. Notify all persons in the area that a spill has
occurred.
2. Prevent the spread of contamination by isolating the
area and covering the spill (absorbent paper).
3. If clothing is contaminated, remove and place in
plastic bag.
4. If an individual is contaminated, rinse contaminated
region with lukewarm water and wash with soap.
5. Notify the radiation safety officer.
6. Wear gloves, disposable lab coat, and booties to clean
up spill with absorbent paper.
7. Put all contaminated absorbent paper in labeled
radioactive waste container.
8. Check the area or contaminated individual with
appropriate radiation survey meter.
officer is notified immediately. The radiation safety offi-
cer typically directs the further response, including
determination of when and how to proceed with
cleanup and decontamination.
In dealing with both minor and major spills, an attempt
is made to keep radiation exposure of patients, hospital
staff, and the environment to a minimum. There are no
absolute guidelines that provide a definitive approach to
every spill. The radiation safety officer must restrict access
to the area until it is safe for patients and personnel.
QUALITY CONTROL IN THE NUCLEAR
PHARMACY
Selected quality control procedures for Tc-99m-labeled
radiopharmaceuticals and for Mo-99/Tc-99m generator
systems are described earlier in this chapter. Considerations
of radiochemical and radionuclidic purity also apply to
other single-photon agents and positron radiopharmaceu-
ticals (Table 1-8). For example, radiochemical purity is
a special concern with radioiodinated agents because of
the potential for uptake of free radioiodine in the thyroid
gland if the radiolabel disassociates from the carrier mole-
cule. Additional quality control procedures in the nuclear
pharmacy are aimed at ensuring the sterility and apyro-
genicity of administered radiopharmaceuticals. Quality
control monitoring of the performance of the dose cali-
brator is also important to ensure that administered doses
are within prescribed amounts.
Sterility and Pyrogen Testing
Sterility implies the absence of living organisms (Table 1-8).
Apyrogenicity implies the absence of metabolic products
Radiopharmaceuticals 17
such as endotoxins. Because many radiopharmaceuticals
are prepared just before use, definitive testing before they
are administered to the patient is impractical, which dou-
bles the need for careful aseptic technique in the nuclear
pharmacy.
Autoclaving is a well-known means of sterilization. It
is useful for sterilizing preparation vials and other uten-
sils and materials but is not useful for any of the radio-
pharmaceuticals employed in clinical practice. When
terminal sterilization is required,various membrane filtra-
tion methods are used. Special filters with pore diame-
ters smaller than microorganisms have been developed
for this purpose. A filter pore size of 0.22 μm is neces-
sary to sterilize a solution. This size traps bacteria,includ-
ing small organisms such as Pseudomonas.
Sterility testing standards have been defined by the
USP. Standard media including thioglycollate and soy-
bean casein digest media are used for different categories
of microorganisms, including aerobic and anaerobic bac-
teria and fungi.
Pyrogens are protein or polysaccharide metabolites
of microorganisms or other contaminating substances
that cause febrile reactions (Table 1-8). They can be
present even in sterile preparations. The typical clini-
cal syndrome is fever, chills, joint pain, and headache
developing minutes to a few hours after injection. The
pyrogenic reaction lasts for several hours and alone is
not fatal.
The USP has established criteria for pyrogen testing.
The historical method involved injecting pharmaceuti-
cal samples into the ear veins of rabbits while measur-
ing their temperature response. The current USP test
uses limulus amebocyte lysate (LAL). The test is based
on the observation that amebocyte lysate preparations
from the blood of horseshoe crabs become opaque in
the presence of pyrogens. The LAL test is more reli-
able, more sensitive, and easier to perform than the
rabbit test.
Radiopharmaceutical Dose Calibrators
The dose calibrator is a key instrument in the radiophar-
macy and is subject to quality control requirements. Four
basic measurements are included: accuracy, linearity, pre-
cision or constancy, and geometry. All of these tests must
be performed at installation and after repair.
Accuracy
Accuracy is measured by using reference standard
sources obtained from the National Institute of Standards
and Technology. The test is performed annually and two
different radioactive sources are used. If the measured
activity in the dose calibrator varies from the standard or
theoretical activity by more than 10%, the device must be
recalibrated.
18 NUCLEAR MEDICINE: THE REQUISITES
Linearity
The linearity test is designed to determine the response
of the calibrator over a range of measured activities.
A common approach is to take a sample of Tc-99m
pertechnetate and sequentially measure it during
radioactive decay. Because the change in activity with
time is a definable physical parameter, any deviation in
the observed assay value indicates equipment malfunc-
tion and nonlinearity. An alternative approach is to use
precalibrated lead attenuators with sequential measure-
ments of the same specimen. This test is performed
quarterly.
Precision or Constancy
The precision or constancy test is designed to measure the
ability of the dose calibrator to repeatedly measure the same
specimen over time. A long-lived standard such as barium-
133 (356 keV, T1/2 10.7 years),cesium-137 (662 keV, T1/2 30
years), or cobalt-57 (122 kev, T1/2 271 days) can be used.
The test is performed daily and observed values should
be within 10% of the value for the reference standard.
Geometry
The geometric test is performed during acceptance test-
ing of the dose calibrator. The issue is that the same
amount of radioactivity contained in different volumes of
sample can result in different measured or observed
radioactivities. For a given dose calibrator, if readings
vary by more than 10% from one volume to another, cor-
rection factors are calculated. For convenience, the cor-
rection factors are based on the most commonly
measured volume of material, which is typically deter-
mined from day-to-day clinical use of the dose calibrator.
Table 1-12 Radiation Doses from Common Diagnostic Nuclear Medicine Procedures
RADIATION DOSIMETRY
Exposure of the patient to radiation limits the amount of
radioactivity that can be administered in the scintigraphic
procedures performed in clinical nuclear medicine. In
general, the exact radiation dose that an individual patient
receives from a nuclear medicine procedure cannot be
calculated. It is not practical to acquire the amount of
data necessary to calculate the actual radiation absorbed
dose for a particular patient. It includes the percent local-
ization of the administered dose in each organ of the
body, the time course of retention in each organ, and the
size and relative distribution of the organs in the body.
Such information is obtained from biodistribution studies
and pharmacokinetic studies in experimental animals
during the development and regulatory approval process
for a new radiopharmaceutical. For each radiopharma-
ceutical, estimates of radiation absorbed doses are made
as part of the approval process and may be taken as “aver-
age”or nominal levels of exposure (Table 1-12).
In brief, the radiation absorbed dose to any organ in
the body depends on biological factors (percent uptake,
biological half-life) and physical factors (amount and
nature of emitted radiations from the radionuclide).
Radiation doses are typically given in rads (radiation
absorbed dose). One rad is equal to the absorption of
100 ergs per gram of tissue. The formula for calculating
the radiation absorbed dose is:
u h S (rk ! rh )
D (rk ! rh ) = A
The formula states that the absorbed dose in a region k
resulting from activity from a source region h is equal to

the cumulative radioactivity given in microcurie-hours in
the source region (Ã) times the mean absorbed dose per
unit of cumulative activity in rads per microcurie-hour
(S ). The cumulative activity is determined from experi-
mental measurements of uptake and retention in the dif-
ferent source regions. The mean absorbed dose per unit
of cumulative activity is based on physical measurements
and is determined by the kind of radiations emanating
from the radionuclide being used.
The total absorbed dose to a region or organ is the sum
of the contributions from all source regions around it and
from activity within the target organ itself. For example,
a calculation of the absorbed dose to the myocardium in
a Tl-201 scan must take into account contributions from
radioactivity localizing in the myocardium and from
radioactivity in the lung, blood, liver, intestines, kidneys,
and general background soft tissues. The percentage
uptake and the biological behavior of Tl-201 are different
in each of those tissues. The amount of radiation reach-
ing the myocardium is also different, depending on the
geometry of the source organ and its distance from
the heart. The formula is applied separately for each
source region, and the individual contributions are
summed.
Factors that affect the dosimetry between patients
include the amount of activity administered originally,
the biodistribution in one patient vs. another, the route
of administration, the rate of elimination, the size of the
patient, and the presence of pathological processes. For
example, for radiopharmaceuticals cleared by the kidney,
radiation exposure is greater in patients with renal fail-
Radiopharmaceuticals 19
ure. Another commonly encountered example is differ-
ing percentage uptakes of radioiodine in the thyroid
depending on whether a patient is hyperthyroid, euthy-
roid, or hypothyroid.
Estimates of radiation-absorbed dose for each major
radiopharmaceutical are provided in tabular form in the
organ system chapters. The tables indicate the absorbed
dose per unit of administered activity for selected organs.
SUGGESTED READING
Chilton HM,Witcofski RL: Nuclear pharmacy: an introduction
to the clinical application of radiopharmaceuticals,
Philadelphia, Lea & Febiger, 1986.
Kowalsky RJ, Perry JR: Radiopharmaceuticals in nuclear med-
icine practice. Norwalk, Conn,Appleton & Lange, 1987.
Ponto JA: The AAPM/RSNA physics tutorial for residents: radio-
pharmaceuticals. Radiographics 18:1395-1404, 1998.
Saha GB: Fundamentals of nuclear pharmacy, 4th ed. New
York, Springer, 1998.
Siegel JA: Guide for diagnostic nuclear medicine and radio-
pharmaceutical therapy. Reston, VA, Society of Nuclear
Medicine, 2004.
Simpkin DJ: The AAPM/RSNA physics tutorial for residents: radi-
ation interactions and internal dosimetry. Radiographics
19:155-167, 1999.
Stabin MG, Breitz HB: Breast milk excretion of radiopharmaceu-
ticals: mechanisms, findings, and radiation dosimetry. J Nucl
Med 41:863-873, 2000.
Swanson DP, Chilton HM,Thrall JH: Pharmaceuticals in med-
ical imaging, New York, Macmillan, 1990.
 2
CHAPTER
Atoms and the Structure of Matter
Bohr Model of the Atom
Electromagnetic Radiation
Mathematics of Electromagnetic Radiation
Relationship of Mass and Energy
Mass Deficit and Nuclear Binding Energy
Radionuclides and their Radiations
Alpha Decay
Beta Minus (β−) or Negatron Decay
Positron Decay and Electron Capture
Isomeric Transition and Internal Conversion
Gamma Ray Emission
Characteristic Radiation and Auger Electrons
Terminology, Units, and Mathematics
of Radioactive Decay
Units of Radioactivity
Half-Life and Decay Constant
Mean Life
Biological Half-Life and Effective Half-Life
Interactions of Radiation with Matter
Negatrons (Beta Particles)
Positrons
Gamma Rays and X-rays
Pair Production
Photoelectric Absorption
Compton Scattering and Compton Effect
Statistics of Radioactive Decay
Medical imaging is based on the interaction of energy with
biological tissues. The kind of diagnostic information avail-
able from each imaging modality is determined by the
nature of these interactions. In conventional x-ray imag-
ing,the differential absorption of x-rays in air,water,fat,and
bone allows the distinction of these tissues in the image.
In ultrasonography,the differing reflective properties of tis-
sues are the basis for creating images. In magnetic reso-
nance imaging, the differences in hydrogen content and in
the chemical and physical environments of hydrogen
nuclei provide the basis for distinguishing tissues.
20
Physics of Nuclear
Medicine
In nuclear medicine,the body is imaged “from the inside
out.” Radiotracers,often in the form of complex radiopharma-
ceuticals,are administered internally. Diagnostic inference
is gained by recording the distribution of the radioactive
material in both time and space. Tracer pharmacokinetics
and selective tissue uptake form the basis of diagnostic util-
ity. To understand nuclear imaging procedures, one must
understand a sequence of concepts, beginning with the
physics of radioactivity, continuing through the process of
detecting radiation and selecting appropriate radiopharma-
ceuticals, and ending with the uptake and distribution of
those pharmaceuticals in health and disease.
ATOMS AND THE STRUCTURE
OF MATTER
Atoms are the building blocks of molecules and are the
smallest structures that represent the physical and chemi-
cal properties of the elements. Each atom consists of a
nucleus surrounded by orbiting electrons (Fig. 2-1). The
nuclei are composed of protons and neutrons,collectively
referred to as nucleons. Protons are positively charged
particles weighing approximately 1.67 × 10−24 g. Their
positive charge is equal in magnitude and opposite to the
charge of an electron (Box 2-1). The element to which
the atom belongs is determined by the number of protons
in the nucleus. Neutrons are slightly heavier than pro-
tons and are electrically neutral, as the name implies.
A shorthand notation has been developed to describe
or define specific atoms. The notation is as follows:
Atomic mass (ZN) Element
A
Z
X N
Atomic number Number
(number of protons) of neutrons
 Physics of Nuclear Medicine 21

where X is the symbol for the element, Z is the number of
protons,N is the number of neutrons,and A is the total num-
ber of neutrons and protons. Z is also referred to as the
atomic number and A as the mass number or atomic mass
number. A nuclide is an atom with a given number of neu-
trons and protons. A radionuclide is simply an unstable
nuclide or nuclear species that undergoes radioactive decay.
Several terms help define special relationships between
different nuclides. The term isotope is used to denote
nuclides with the same number of protons (Z), that is, the
same element but different numbers of neutrons (N). For
example, the element iodine has more than 20 isotopes.
All except one (I-127) are radioisotopes or radionuclides
and several are of medical interest, including I-123, I-125,
and I-131,which have the following notations:
Electrons
Nucleus
Figure 2-1 Bohr model of the atom. The nucleus contains
protons and neutrons and has a radius of 10−14 m. The protons in
the nucleus carry a positive charge. The orbital electrons carry a
negative charge.
123 125 131
I I I
35 70 35 72 35 78
Other special terms are: isobar, to indicate the same A
but different N and Z; isotone, to indicate the same num-
ber of N but different Z and A; and isomer, to indicate dif-
ferent energy states in nuclides with identical A, Z, and N.
The most important isomers in nuclear medicine are
technetium-99 and technetium-99m, in which the m
denotes a metastable or prolonged intermediate state in
the decay of molybdenum-99 to technetium-99.
BOHR MODEL OF THE ATOM
In the classic Bohr model of the atom, electrons are
arranged in well-defined orbits around the nucleus (Figs.2-1
and 2-2). The number of orbital electrons in each atom
equals the atomic number,Z (the number of protons in the
nucleus). The closest orbit, referred to as the K shell, is fol-
lowed by the L, M, and N shells and so forth. The maximum
number of electrons in the K shell is 2, in the L shell is 8, in
the M shell is 18, and in the N shell is 32, except that no
more than 8 electrons can be in the outermost shell of an
atom. Figure 2-2 is a simplified schematic of the Bohr
model for potassium. The term valence electron is used to
designate electrons in the outermost shell (Box 2-2). These
electrons are important in defining the chemical properties
of elements. For example, atoms with the outermost shell
maximally filled are chemically unreactive (e.g., the inert
gases helium,neon,argon,krypton,xenon,and radon).
Electrons have a negative charge equal to 1.6 × 10−19
coulomb; as previously noted, protons have a positive
charge of equal magnitude. Electrons are bound in their

Box 2-1 Summary of Physical Constants

Speed of light in a vacuum (c) 3.0 × 108 m/sec

Elementary charge (e) 4.803 × 10−10 esu
1.602 × 10−19 coulomb
Rest mass of electron 9.11 × 10−28 g
Rest mass of proton 1.67 × 10−24 g N M L K Nucleus
Planck’s constant (h) 6.63 × 10−27 erg sec Z = 19

Avogadro’s number 6.02 × 1023 molecules

gram mole
1 electron volt (eV) 1.602 × 10−12 erg
1 calorie (cal) 4.18 × 107 erg
1 Angstrom (Å) 10−10 m
Euler’s number (e) (base 2.718
of natural logarithms)
Atomic mass unit (U) 1.66 × 10−24 g (12⁄ the
mass of a carbon-12
atom) Figure 2-2 Potassium atom. Potassium has an atomic number
of 19, with 19 protons in the nucleus and 19 orbital electrons.
22 NUCLEAR MEDICINE: THE REQUISITES
orbits by the electrical force between their negative
charge and the positive charge of the nucleus. The high-
est binding energy is in the electrons in the shell closest
to the nucleus (K shell),with progressively lower binding
energies in the more distant shells. Before an electron
can be removed from its shell, the binding energy must
be overcome. Interactions involving orbital electrons
and ionizing electromagnetic radiation (x-rays and
gamma rays) are central to the way medical images are
made and to the quality of the images.
Box 2-2 Terms Used to Describe Electrons
Term Comment
Electron Basic elementary particle
Orbital electron Electron in one of the shells or
orbits in an atom
Valence electron Electron in the outermost shell of
an atom; responsible for
chemical characteristics and
reactivity
Auger electron Electron ejected from an atomic
orbit by energy released during
an electron transition
Photoelectron Electron ejected from an atomic
orbit as a consequence of an
interaction with a photon
(photoelectric interaction) and
complete absorption of the
photon’s energy
Conversion Electron ejected from an atomic
electron orbit because of internal
conversion phenomenon as
energy is given off by an
unstable nucleus
Photon energy 106 103
electron Volt (eV) 1MeV 1keV
0.0001 nm 0.01 nm
Gamma rays X-rays
Figure 2-3 Electromagnetic energy spectrum. Photon energies (eV) and wavelengths of gamma
and x-rays, ultraviolet, visible light, infrared, and radiowaves.
It has long been recognized that the Bohr model of
the atom is too simplistic to portray many atomic phe-
nomena accurately. Nuclear physicists have developed
sophisticated wave mechanical or quantum mechanical
models in which probability density functions are used
to describe spatial and temporal properties of electrons.
However, the Bohr model can still be used to describe
the basic interactions of interest in nuclear medicine.
ELECTROMAGNETIC RADIATION
The term electromagnetic radiation or electromagnetic
waves refers to energy in the form of oscillating electric
and magnetic fields. Individual packets of electromagnetic
radiation are referred to as photons. Photons with energy
greater than 100 eV are classified as x-rays or gamma rays.
Lower energy photons may be in the range of ultraviolet
light, infrared, visible light, radar waves, or radio and televi-
sion waves (Fig. 2-3). The unit of energy used to describe
these electromagnetic waves or radiations is the electron
volt (eV). One electron volt is defined as the kinetic
energy of an electron accelerated through a potential differ-
ence of 1 volt (1 eV = 1.6 × 10−19 joules or 1.6 × 10−12 erg).
Mathematics of Electromagnetic Radiation
The relationship between the energy of x-rays and gamma
rays (or other electromagnetic radiations) and their frequen-
cies is given by E = hv, where v is the frequency and h is
Planck’s constant (see Box 2-1). Electromagnetic radiation
travels with the speed of light (c). The relationship between
frequency and wavelength is given by c = vλ,where λ is the
wavelength. Rearranging this equation to solve for v and
substituting it into the previous equation yields:
E = hc
m
100 101 106 1010
1eV
Increasing Wavelength
10 nm 1000 nm 0.01 cm 1 cm 1m 100 m
Ultra-
Infrared Radio waves
violet
Radar TV FM AM
Visible light

Taking wavelength in angstroms (Å) and energy in keV and
substituting the numerical value for h and c, this becomes:
E (keV) = 12.4
m (Ac )
RELATIONSHIP OF MASS
AND ENERGY
In 1905,Albert Einstein published his famous equation
E = mc2, where E is energy in ergs, m is mass in grams,
and c is the velocity of light in a vacuum (3 × 10 8
m/sec). From this equation, it is possible to calculate
the energy equivalent of the various subatomic particles.
By definition the unified or universal atomic mass unit
(U) is equal to one-twelfth the mass of a carbon-12 atom
(1 U = 1.66 × 10−24 g) (Box 2-1). Using this value for
mass in Einstein’s equation yields the following result:
E = (1.66 × 10−24 g/U) × (3.0 × 1010 cm/sec)2
E = 1.5 × 10−3 erg/U
(1 erg = 1 gcm2/sec2)
Inserting the conversion factor between ergs and electron
volts (Box 2-1) yields the relationship 1 U = 931.5 MeV.
Table 2-1 provides the mass and energy relationships for
the basic subatomic particles. The most important of
these relationships in clinical nuclear medicine and
positron emission tomography is the energy equivalence
of the mass of an electron, which is 511 keV.
Mass Deficit and Nuclear Binding Energy
The relationships between mass and energy are of funda-
mental importance in nuclear physics. By carefully
determining the weight of atomic nuclei, physicists have
shown that the theoretical sum of the component nucle-
ons is always greater than the actual observed mass of
the respective atomic nuclei. The difference is known as
the mass deficit. The nuclear binding energy is defined
as the energy equivalent of the mass deficit.
Energy equal to the difference in nuclear binding
energy of the pretransformation and posttransformation
nuclei is released in atomic fusion and atomic fission.
The energy of hydrogen and atomic bombs comes from
energy released when trillions of new atomic nuclei are
Table 2-1 Mass-Energy Equivalence for Atomic
Particles
Particle Mass (U) Energy (MeV)
Electron 5.486 × 10−4 0.511
Proton 1.0073 938.20
Neutron 1.0087 939.5
Physics of Nuclear Medicine 23
formed. That is, the aggregate mass deficit of the post-
transformation nuclei after a fusion or fission reaction is
greater than that of the original nuclei.
The concept of mass deficit is also fundamental to the
use of radionuclides in medical imaging. As a more sta-
ble atomic configuration is formed in the radioactive
decay process, the mass deficit always increases. In
many radionuclide decay schemes, part of the mass
deficit is given off in the form of energetic electromag-
netic radiation (photons) that can be detected and used
to form medical images.
RADIONUCLIDES AND THEIR
RADIATIONS
Because of their physical properties, certain atoms are
unstable and undergo radioactive decay. The daughter
product in radioactive decay is always at a lower energy
state than the parent. The energy difference or mass
deficit between parent and daughter is equal to the total
energy in the radiations emitted. For each radionuclide,
the type of radiation given off, the energy of the radia-
tion(s), and the half-life of the decay process are physical
constants. These parameters are important in determin-
ing the suitability of a given radionuclide for medical use.
The types of radiation important in nuclear medicine
are gamma rays, characteristic x-rays, negatrons (beta par-
ticles), positrons (beta particles), and alpha particles. By
definition, the term gamma ray is used for photons orig-
inating in the nucleus and the term x-ray for photons
originating outside the nucleus.
Among the lighter atomic elements, the number of
protons and neutrons in the nucleus is roughly equal.
As the atomic number Z increases, the ratio of neutrons
to protons in stable nuclei increases. A plot of this ratio
vs. atomic number defines an empirical “line of stabil-
ity” (Fig. 2-4). That is, the neutron/proton (N/P) ratio is
greater than 1 for stable nuclei in the middle and upper
atomic numbers. This observation is important in pre-
dicting the mode of radioactive decay of unstable
nuclides. In general, the decay process tends to return
the daughter nucleus closer to the line of stability. That
is, if an unstable nucleus contains more neutrons than do
stable isotopes of the same element, the mode of decay
will reduce the N/P ratio, and vice versa for nuclei with
fewer neutrons than predicted by the line of stability.
A system of schematic diagrams has been developed
to illustrate radioactive decay. Positive emissions (alpha
particles and positrons) and electron capture cause the
daughter nucleus to have a lower atomic number, which
is indicated by an arrow pointing down and to the left
(Fig. 2-5). Following negative emissions by beta minus
particles (negatrons), the daughter nucleus has a higher
atomic number, which is indicated by an arrow pointing
down and to the right (Fig. 2-6).
24 NUCLEAR MEDICINE: THE REQUISITES

160 Complete decay schemes can be complex, with multi-

Neutron rich ple pathways from parent to daughter. For practical pur-
(Beta minus decay) poses, the decay schemes in this book are simplified to
140 Proton rich
illustrate important general principles and specific
(Electron aspects relevant to clinical nuclear medicine.
capture
120 and
positron
decay) Alpha Decay
100
Alpha particles are essentially helium nuclei (i.e.,two pro-

”
tons and two neutrons) with a +2 charge and an atomic
ility
Neutron number N

tab
80 mass number of 4. Alpha decay is common in the higher
f “s

atomic number range of the periodic table of elements.

eo

N
For example, radium-226 (Ra-226) decays to radon-222
Lin

60 Z

(Rn-222) by emitting an alpha particle (Fig. 2-5).
In the simplified scheme shown for Ra-226, three dif-
40 ferent alpha particles are shown (Fig. 2-5). One reaches
the ground state of Rn-222 directly. The other two result
20
in an excited state of Rn-222 with subsequent gamma ray
emission to reach the ground state. In the complete
decay scheme for Ra-226, additional alpha particles are
0
present but they occur in low abundance.
20 40 60 80 100 In all radioactive decay processes,mass and energy are
conserved. The transition energy is the total energy
Atomic number Z
released during the decay process. For alpha decay, this
Figure 2-4 Ratio of neutrons to protons. For low atomic energy is in the form of the kinetic energy of the alpha
number elements, the two are roughly equal (Z = N). With
increasing atomic number, the relative number of neutrons particle plus the energy released in the form of gamma
increases. Stable nuclear species tend to occur along the line of radiation.
“stability.”

226
Ra (1600 yr)
88

448 keV


2
186 keV


1

222
Rn (3.8 days)
86
Figure 2-5 Decay scheme for radium-226. Decay is by alpha particle emission to the daughter
product radon-222. The emission of an alpha particle results in a decrease in atomic number of 2
and a decrease in atomic mass of 4.

131
I (8 days)
53

364 keV
 (81%)
131
Xe (stable)
54
Figure 2-6 Decay scheme for iodine-131. Decay is by negatron
emission. In negatron or beta minus decay the atomic mass does
not change (isobaric transition). The atomic number increases by
one. The daughter, xenon-131, has one more proton in the
nucleus.
Alpha particles are undesirable in diagnostic applica-
tions because their range in soft tissue is a fraction of a
millimeter. Therefore, they do not contribute to the
image and result in high radiation to the patient without
any contribution of diagnostic information. No currently
used diagnostic radiopharmaceuticals include alpha-
emitting radionuclides. On the other hand, a number of
therapeutic agents have been designed to incorporate
alpha particle emitters.
Beta Minus (b-) or Negatron Decay
The negatron decay process involves the conversion of a
neutron into a proton, an electron, and a subatomic parti-
cle called an antineutrino. The electron is ejected from
the atomic nucleus, thereby giving the decay process its
name. The term negatron is used to distinguish negative
electrons from positive electrons, or positrons. Negatrons
and positrons are also referred to as “beta particles.”
Depending on their charge, they can also be referred to as
beta minus (β−) or beta plus (β+) particles. Thus, nega-
tron decay is also called beta decay. Negatrons or beta
minus particles are identical to orbital electrons in both
mass and charge, differing only in the fact that they origi-
nate from the nucleus of the atom.
The N/P ratio decreases as a result of negatron decay.
This mode of decay could be predicted to occur in neu-
tron-rich nuclei; that is, it occurs in nuclei with more neu-
trons than the stable species in the respective part of
the atomic chart. For example, stable iodine has a mass
number of 127 (53 protons, 74 neutrons). By compari-
son, I-131 has 78 neutrons. Because 78 neutrons is a
higher number than stable iodine, I-131 undergoes beta
minus decay (Fig. 2-6).
The transition energy in negatron decay is given off in
the form of kinetic energy of the beta particle, the energy
Physics of Nuclear Medicine 25
in the antineutrino, and the energy in any associated
gamma radiation. The maximum kinetic energy (Emax)
that a beta particle can have is a physical constant of the
decay process. Beta particles are emitted with a continu-
ous spectrum of energies lower than the maximum. The
mean kinetic energy of beta particles (Eb) is approxi-
mately one third of the maximum (Eb = 13⁄ Emax). For beta
particles with less than the maximum kinetic energy, the
energy is shared between the beta particle and the anti-
neutrino.
Again, a decay scheme can have more than one path-
way from the parent to the daughter. For many radionu-
clides decaying by negatron decay, beta particles with
different maximum kinetic energies are emitted.
Because the total transition energy must be the same for
each pathway, the energy of associated gamma radiation
is also correspondingly different. For example, the
decay scheme for I-131 presented in Fig. 2-6 illustrates
only one pathway from parent to daughter, the one of
most interest and importance in clinical practice. In
reality, there are beta particles given off with six differ-
ent energies and there are 19 different gamma rays.
However, the most abundant gamma ray, with an energy
of 364 keV, occurs in 81% of transitions (known as an
abundance of 81%).
A number of beta-emitting radionuclides have been
used in clinical nuclear medicine. I-131,the first radionu-
clide of importance in medicine, is still used. The disad-
vantage of beta emitters is the high radiation dose
received by the patient from the beta particles. For
radioiodine-131, this disadvantage becomes an advantage
when the radionuclide is used as therapy for thyroid can-
cer and hyperthyroidism.
Positron Decay and Electron Capture
As the name implies, in positron decay a positive elec-
tron or positively charged beta particle is ejected from
the nucleus, resulting in a decrease in the atomic number
between the parent and the daughter nuclei and an
increase in the N/P ratio. Positron decay occurs in
nuclides that are neutron poor, with N/P ratios less than
those occurring on the line of stability. Positron decay is
illustrated in Fig. 2-7 for fluorine-18. The minimum tran-
sition energy required for positron decay is 1.02 MeV,
which is the energy equivalent of the mass of two elec-
trons. The transition energy in excess of the 1.02 MeV
threshold is embodied in the kinetic energy of the
positrons and any associated gamma rays. For positrons
given off with less than maximum kinetic energy, the
energy difference is in subatomic particles called neutri-
nos. In both positron and negatron decay, the neutrinos
(or antineutrinos) carry away a substantial portion of the
transition energy. The likelihood of neutrinos reacting in
soft tissue is small and the energy in neutrinos is not
26 NUCLEAR MEDICINE: THE REQUISITES
important in calculating radiation dosimetry for clinical
applications.
In unstable nuclei where the maximum available tran-
sition energy is less than 1.02 MeV,decay of neutron-poor
radionuclides is by electron capture (Fig. 2-8). In elec-
tron capture, an electron from one of the orbital shells
close to the nucleus is incorporated into the nucleus,
converting a proton into a neutron. The captured elec-
tron is usually from the K shell. The resulting vacancy is
filled by transition of an electron from a shell farther
from the nucleus. The energy released from this elec-
tron transition appears either as characteristic x-radia-
tion or as the kinetic energy of an Auger electron.
Some radionuclides decay by multiple modes, includ-
ing electron capture, positron decay, and negatron decay
(Fig. 2-9). The likelihood of electron capture increases
as the available transition energy decreases. The proba-
bility of electron capture also increases with increasing
atomic number.
18
9 F (110 min)

18
8O (stable)
Figure 2-7 Decay scheme of fluorine-18 by positron emission.
The daughter product, oxygen-18, has one fewer proton in the
nucleus. Positron decay is another example of an isobaric
transition without change in atomic mass between parent and
daughter.
201
81Tl (73 hr)
EC
167 keV
 (3%)
 (10%) 32 keV
201
80 Hg (stable)
Figure 2-8 Thallium-201 decay by electron capture (EC). The
daughter nucleus (mercury-201) has one fewer proton than the
parent.
Isomeric Transition
and Internal Conversion
No radionuclide undergoes true radioactive decay just
by the emission of gamma radiation. However, in some
decay schemes, there are intermediate species with
measurable half-lives that exist in a metastable state.
The concept of metastability is arbitrary. Most gamma
rays are emitted almost immediately (10−12 seconds)
after the primary decay process, whether it be alpha
decay, negatron decay, positron decay, or electron cap-
ture. When the intermediate excited state lasts longer
than 10−9 seconds, the term metastable is used and an m
is placed after the mass number to indicate the phenom-
enon. The transition from the metastable state to the
ground state is isomeric because the atomic number
does not change and thus the transformation is from one
isomer to another.
The most important example of a metastable state in
nuclear medicine practice is technetium-99m (Tc-99m),
which occurs in the decay of molybdenum-99 to tech-
netium-99 (Figs. 2-10 and 2-11). The metastable state for
Tc-99m has a half-life of 6 hours which allows ample time
for the separation of the metastable species from the par-
ent radionuclide and its subsequent use for clinical imag-
ing procedures. Tc-99m is attractive from a radiation
safety or health physics standpoint because it is essen-
tially a pure gamma emitter not associated with primary
126
53 I
EC,  
126 126
52Te (stable) 54Xe (stable)
Figure 2-9 Iodine-126 decay through multiple processes. The
diagram indicates decay by electron capture and by the emission
of both positrons and negatrons.
99
42Mo (2.8 days or 66 hr)

142.7 keV
 140.5 keV

0.0
99
43Tc (2.1  10 yr)
5
Figure 2-10 Decay scheme of molybdenum-99. Negatron
emission to technetium-99.
 Physics of Nuclear Medicine 27

particulate radiations. Its use as a radiolabel is associated
with favorably low radiation dosimetry.
The energy released in isomeric transitions may be
used to dislodge an orbital electron instead of being emit-
ted as a gamma ray. This process is called internal con-
version (Fig. 2-12). The kinetic energy of the electron
(conversion electron) is equal to the difference between
the gamma ray energy and the binding energy of the
electron. The internal conversion process reduces the
number of usable, detectable gamma photons for imag-
99m
43 Tc (6.01 hr)
 140.5 keV (89%)
99
43 Tc
Figure 2-11 Isomeric transition of technetium-99m to
technetium-99.
ing. It also results in a higher radiation dose to the patient
because the conversion electron is absorbed in tissue
close to its site of origin. In the “decay” of Tc-99m, a 140-
keV gamma ray is given off 89% of the time and internal
conversion accounts for most of the remaining transitions.
Gamma Ray Emission
Many radioactive decay processes result in the release of
gamma rays or gamma photons, which are ionizing elec-
tromagnetic radiations that originate in the excited,
unstable atomic nucleus. They have discrete energies
defined by the decay scheme for the respective radionu-
clide and occur over a wide range of energies. Gamma
rays most useful in conventional single-photon nuclear
medicine applications have energies of approximately
80-400 keV. Nuclear medicine imaging equipment has
been optimized for this energy range. Photons with
energies less than 80 keV present difficulties because of
their relatively high attenuation in tissue and their scat-
tering properties. They are also less reliably localized by
standard imaging devices because of the smaller amount
of total energy available in the detection process.
Gamma rays with energies significantly greater than 400
keV are progressively more difficult to image with con-
ventional gamma cameras. The detection efficiency in
gamma camera systems is less at higher energies. Spatial

Nucleus Nucleus

 Ray
from nucleus
Vacancy

Vacancy Electron transition

between shells

Conversion Characteristic
A electron B x-ray
Figure 2-12 Internal conversion and characteristic x-ray emission. A, With internal conversion,
an orbital electron is ejected from its shell, instead of the emission of a gamma ray. B, A
characteristic x-ray is then given off as a consequence of the electron vacancy’s being filled.
28 NUCLEAR MEDICINE: THE REQUISITES

resolution is also lost through difficulty in collimating
high-energy photons.
Characteristic Radiation
and Auger Electrons
When an orbital electron is removed from its shell, it
leaves a vacancy that is rapidly filled by a free electron or
an electron from a shell farther from the nucleus. In this
process,the “cascading”electron gives up energy as it fills
in the vacancy and becomes more tightly bound. Most
often, the energy that is given up by the electron is emit-
ted in the form of electromagnetic radiation.
The electromagnetic radiations that arise in the
process of filling a vacancy are called characteristic radi-
ations or characteristic x-rays where applicable because
their energy is uniquely defined by the difference in the
binding energy of the donor shell and the shell where
the vacancy is being filled (that is, the x-ray energy
is “characteristic” of the respective transition) (see Fig.
2-12). In some applications of radionuclides, detection
of characteristic x-rays is the primary means of forming
the image or measuring the amount of radioactivity. An
example of this is myocardial perfusion imaging with
thallium-201 (Tl-201). The most abundant photons used
for imaging are actually characteristic x-rays from mer-
cury-201, the daughter product of Tl-201 decay.
An alternative process to the emission of characteris-
tic radiation is the ejection of another electron by the
energy released in filling a given vacancy. An electron
ejected in this way is termed an Auger electron (Box 2-2).
The probability of the emission of characteristic or
fluorescent x-rays relative to Auger electrons is higher for
more tightly bound electrons. Therefore, the emission of
characteristic x-rays is more common for nuclei with
a higher Z number and for electrons in the inner shells.
An alternative to the curie in the international system
(SI) of units is the becquerel (Bq), which is equal to
1 dps. The relationship between the curie and the bec-
querel is straightforward, but may be somewhat confus-
ing to those used to the older term. One millicurie
equals 37 million Bq, or 37 MBq. Both terminology sys-
tems are used widely in the literature (Table 2-2).
However, the SI system is increasingly preferred.
Half-Life and Decay Constant
The mathematics of radioactive decay follow from direct
physical measurements. The fundamental empirical
observation determined early in the history of work with
radionuclides is that the number of atoms undergoing
decay during any finite period of time is proportional to
Box 2-3 Conversion of International
System (SI) and Conventional
Units of Radioactivity
CONVENTIONAL UNIT
1 curie (Ci) = 3.7 × 1010 disintegrations per second (dps)
SI UNIT
1 becquerel (Bq) = 1 dps
CURIES → BECQUERELS
1 Ci = 3.7 × 1010 dps = 37 GBq
1 mCi = 3.7 × 107 dps = 37 MBq
1mCi = 3.7 × 104 dps = 37 KBq
BECQUERELS → CURIES
1 Bq = 1 dps = 2.7 × 10−11 Ci = 27 pCi
1 MBq = 106 dps = 2.7 × 10−5 Ci = 0.027 mCi
1 GBq = 109 dps = 27 mCi

TERMINOLOGY, UNITS,
AND MATHEMATICS
Table 2-2 Conversion from Centimeter-gram-
OF RADIOACTIVE DECAY second (CGS) System to International
System (SI) Units
Units of Radioactivity
Two systems for expressing decay or disintegration rates Conversion
are in widespread use and are potentially confusing. The CGS unit SI unit factor
more widely-used system historically was based on
Work erg joule (J) 107
the curie. This unit was based on the disintegration rate
Radioactivity curie (Ci) becquerel (Bq) 3.7 × 1010
of 1 gram of radium and was defined as 3.7 × 1010 disinte- Radiation rad gray (Gy) 100
grations per second (dps) (Box 2-3). It is now known absorbed
that the disintegration rate of 1 gram of radium is slightly dose
different than 1 curie, but the quantitative definition has Radiation roentgen (R) coulomb/kg 2.58 × 10−4
exposure
been widely used throughout the world. Most medical
Roentgen rem sievert (Sv) 100
diagnostic applications involve amounts of radioactiv- equivalent
ity in the microcurie (3.7 × 104 dps) or millicurie (3.7 × man
107 dps) range.
 Physics of Nuclear Medicine 29

the number of radioactive atoms in the sample. This pro- graph paper, the function is a straight line with a slope
portion can be written: equal to −λ (Fig. 2-13).
- dN t From the preceding fundamental equations, it is possi-
dt aN t
ble to derive the concept of physical half-life, which turns
where Nt is the number of radioactive atoms in the sam- out to be a more intuitive and useful way of describing
ple at time t. The term dNt/dt is mathematical notation radioactive decay than using the decay constant. The half-
expressing the change in the number of radioactive life is simply defined as the time required for the number
atoms over a short interval. The negative sign in the of radioactive atoms in a sample to decrease by exactly
equation denotes that the number of radioactive atoms one-half or 50%. Mathematically, the value of the half-life
decreases over time. can be derived from the previous equations by substitut-
For any given radioactive species, the equation may be ing N/2 and T on the two sides respectively as follows:
1
2
rewritten as: N0 - mT
- dN t 2 = N0 e
1
2

dt mN t 1 = e- m T 1
2

The term λ is the constant of proportionality and is a 2

mathematical constant for each radionuclide. It is also Because e - 0.693 =1/2,this equation can be simplified to
called the decay constant and has units of 1/time. yield:
The last equation can be rearranged and integrated
λT = 0.693
1
and provides the classic equation: 2

0.693
T =
m
1

N t = N 0 e- mt
2

From the preceding equations, the decay constant has

The term N0 represents the number of radioactive atoms units of reciprocal time (e.g., sec−1 or hr−1) and the physi-
at time t = 0 and e is Euler’s number, which equals ~2.718 cal half-life has units of time such as seconds, hours, days,
(Box 2-1). In words, this equation says that the number or years. Radionuclides with long physical half-lives have
of radioactive atoms at any later point in time (Nt) is smaller values for the decay constant. That is, the longer
equal to the product of the original number (N0) times an the physical half-life, the smaller the fraction of the
exponential factor (e) that takes into account the rate of radioactive atoms that undergoes disintegration in any
decay (l) and the length of time after the initial measure- given unit of time. From a practical standpoint, most
ment (t). Because the activity of the sample is propor- radionuclides used in clinical nuclear medicine must have
tional to the number of atoms in that sample (A = λN), half-lives of hours or days, which permits shipping from
the equation can be rewritten as: the manufacturing site to the hospital, preparation of the
At = A0 e- mt radiopharmaceutical, and imaging. Use of shorter-lived
agents is feasible in institutions with radionuclide produc-
where A indicates activity in either curies or becquerels. tion facilities such as cyclotrons or special accelerators.
The decay curve plotted on standard coordinates with In certain cases, radionuclides are obtained from “gen-
time on the x-axis and activity on the y-axis for a radioac- erator” systems, and the practical limitation is then the
tive sample shows an exponentially decreasing function half-life of the parent compound. For example, the half-
that approaches but never reaches zero. On semilog life of Tc-99m is 6 hours. The half-life of its parent,

1 1.0
Radioactivity (fraction remaining)

Radioactivity (fraction remaining)

0.9
0.8
0.7
0.6
0.5 0.1
0.4
0.3
0.2
0.1
0 0.01
0 6 12 18 24 30 36 42 0 6 12 18 24 30 36 42
Time (hr) Time (hr)
A B
Figure 2-13 Decay plot for technetium-99m. A, standard and, B, semilog graphs.
30 NUCLEAR MEDICINE: THE REQUISITES

molybdenum-99 (Mo-99), is 2.7 days (see Figs. 2-10 and

2-11). The Mo-99/Tc-99m generator system provides INTERACTIONS OF RADIATION
the dual advantage of a longer-lived parent, which per- WITH MATTER
mits commercial distribution and prolonged on-site
availability, while the short half-life of the Tc-99m daugh- Negatrons (Beta Particles)
ter reduces radiation exposure to the patient compared Negatrons,or beta particles,cause ionization in tissues by
with longer-lived agents. electrostatic interactions with orbital electrons. They
give up energy through a series of such interactions
along a tortuous path. As a rule of thumb, the maximum
Mean Life penetration of beta particles in soft tissue in centimeters
The concept of the mean life of a radionuclide is useful is equal to the maximum kinetic energy of the negatron
in thinking about radiation dosimetry. The mean life is in megaelectron volts divided by two. Thus, the radia-
given as: tion dose delivered by negatrons in soft tissue is rela-
tively close to their source. For example, the maximum
t= 1 kinetic energy of the most abundant beta particle in the
m
or decay of I-131 is 0.606 MeV. The majority of the radia-
tion dose delivered in I-131 therapy is within 0.3 cm of
t = 1.44 T 1
2
the location of the nucleus undergoing decay.
The concept of mean life is more difficult to understand
intuitively than the concept of physical half-life but may Positrons
be thought of as the average length of time of the radioac-
tive atoms in a sample before they undergo disintegration. Positrons also give up their kinetic energy through elec-
trostatic ionizations. As the positron approaches ther-
mal energy, it undergoes annihilation by combining
Biological Half-Life and Effective Half-Life with a negatively charged electron (Fig. 2-14). Two
An important concept in determining radiation exposure gamma photons are emitted 180 degrees apart. Each has
to patients is the biological half-life and the corollary an energy of 0.511 MeV, the energy equivalent of
concept, effective half-life. The term biological half-life positron-electron mass. This unique phenomenon of
is used to describe the biological clearance of the annihilation radiation 180 degrees apart is the basis for
radionuclide from a particular tissue or organ system. positron emission tomography (PET).
Thus the actual half-life or effective half-life of a radio-
pharmaceutical in a biological system is dependent on Gamma Rays and X-Rays
both the physical (p) half-life and the biological (b) clear-
ance. Because physical decay and biological clearance Gamma rays and x-rays are attenuated in tissues through
occur simultaneously in parallel, the relationship three processes. Photons can be completely absorbed
between them and the effective half-life is given by: by the photoelectric effect or in pair production. They
can also undergo scattering or deflection from their orig-
1 = 1 + 1 inal path by the Compton effect or Compton-scattering
T 1
2 eff T p T b
1
2
1
2 phenomenon, in which photons give up part of their
original energy.
Rearranging terms, this becomes:
T
1
b #T 1
p
=
2 2
T
+T
1
2 eff T
1 1
2 b 2 p

The concept of biological half-life is not as mathemati-

cally clear-cut as the physical half-life. It can vary among
subjects and does not necessarily follow a regular expo- Electron (e) Positron (e)
nential process. For example, the biological half-life of
radioactivity in the bladder is determined by the time at
which a patient chooses to void. The half-life of xenon-
133 in the lung during pulmonary ventilation studies is 511 keV 511 keV
determined by the rate and depth of respiration and by 180
 
the presence of pulmonary disease. Nonetheless, the
Figure 2-14 Positron annihilation. The mass of a positron and
term biological half-life is useful in thinking about the an electron is converted to energy in the form of two photons.
amount of exposure the patient actually receives during The photons each have an energy of 0.511 MeV and are given off
a nuclear medicine procedure. 180 degrees apart.
 Physics of Nuclear Medicine 31

Pair Production
Pair production requires a photon with a minimum
energy of 1.02 MeV. The photon energy is converted
into one negative and one positive electron. Because the
energy required is greater than the photon energies used
in medical imaging, this form of attenuation is not impor-
tant in nuclear medicine.
Photoelectric Absorption
Photoelectric absorption occurs when the total energy of
an x-ray or gamma ray photon is transferred to an orbital
electron (Fig. 2-15, A). The photon must possess energy
greater than the binding energy of the electron. The elec-
tron is displaced from its orbit or shell and is either lifted
to a higher shell or ejected from the atom (Fig. 2-15, B).
Ejected electrons are termed photoelectrons.
As a consequence of the photoelectric interaction, an
electron cascade occurs to fill the vacancy, with the subse-
quent emission of characteristic x-rays or Auger electrons
(Fig. 2-15, C). Photoelectric absorption is most likely to
occur when the photon energy is just above the electron
binding energy. The kinetic energy of the photoelectron
is equal to the difference between the energy of the inci-
dent photon and the electron binding energy.
For a given absorbing material, as photon energy
increases,the likelihood of a photoelectric event decreases.

Photoelectron

Nucleus

Nucleus

Vacancy

Incident photon
A B
Characteristic
x-ray

Vacancy
Nucleus
Electron
transition
between
shells

C
Figure 2-15 Photoelectric absorption. A, An incident photon interacts with an orbital electron.
B, The electron is ejected from its shell, creating a vacancy. The electron is either ejected from the
atom or moved to a shell farther from the nucleus. C, The orbital vacancy is filled by the transition
of an electron from a more distant shell. A characteristic x-ray is given off as a consequence of this
transition.
32 NUCLEAR MEDICINE: THE REQUISITES
The photoelectric interaction is important in soft tissues
up to an energy of approximately 50 keV. Radionuclides
with associated photon energies lower than 50 keV are
less desirable for clinical applications because of the high
absorption of these photons in soft tissue owing to pho-
toelectric interaction.
Although photoelectric absorption is undesirable in
body tissues, it is fundamental to the detection of ionizing
radiation. In both nuclear medicine and roentgenogra-
phy,the creation of images depends on energy absorption
in a detecting medium through the photoelectric interac-
tion. For this reason, imaging systems typically are high-
density, high-Z materials such as inorganic crystals, in
which the likelihood of photoelectric absorption is high.
Compton Scattering and Compton Effect
In Compton scattering, a photon interacts with a weakly
bound outer shell electron. Instead of being completely
absorbed as in the photoelectric interaction, in the
Compton process the photon is deflected from its origi-
nal direction and continues to exist but at lower energy
(Fig. 2-16). The energy difference is transferred to the
recoil electron as kinetic energy. Compton scattering is
the dominant mode of gamma ray and x-ray interaction in
soft tissues between 30 keV and 30 MeV.
Because the Compton-scattered photon gives up
energy in the interaction, its wavelength increases. The
formula for this is:
Λl = 0.0243 (1 − cos f)
Nucleus Compton
electron
Angle
Compton-
of scatter
scattered
photon
Incident
photon
Figure 2-16 Compton scatter.An incident photon interacts
with an outer or loosely bound electron. The photon gives up a
portion of its energy to the electron and undergoes a change in
direction at a lower energy.
where Ll is the change in wavelength and the angle f is
the angle through which the photon is scattered. The
angle of scatter can be minimal or up to 180 degrees
(backscatter).
The significance of Compton scattering in nuclear
imaging is that scattered photons reaching the imaging
detector must be discriminated against and not allowed
to form part of the image. Because Compton-scattered
photons give up part of their energy, one way to discrimi-
nate against them is through setting an “energy window”
for acceptance of events in the detector. However, pho-
tons scattered through a relatively narrow angle lose only
small amounts of energy and may not be effectively
excluded by pulse height analysis and the setting of an
energy window. Thus, Compton-scattered photons con-
tribute to the loss of spatial resolution in nuclear medi-
cine images. The problem is progressively worse for
lower energies because the lower the original photon
energy,the less the change in energy for a given scattering
angle.
STATISTICS OF RADIOACTIVE DECAY
The time of decay of any single unstable radioactive
nucleus is unpredictable and is not influenced by the
decay of other nuclei or the physical or chemical envi-
ronment of the nucleus. Because radioactive decay is
random, the actual observed number of nuclei undergo-
ing decay in any given period is subject to statistical
uncertainty; this uncertainty is a practical problem in
clinical nuclear medicine. In any setting where a quantita-
tive measurement is required, such as determining the
amount of radioactivity in a radiopharmaceutical to be
given to the patient or in a blood sample used in calculat-
ing a physiologic parameter or performing quality con-
trol of nuclear instrumentation, estimates of statistical
certainty are necessary.
Radioactive decay follows Poisson statistics or the
Poisson probability law. The Poisson probability density
function is similar but not identical to the Gaussian or
normal probability density function. Curves expressing
the Poisson and Gaussian probability density functions
are more closely matched as the number of observed
events is increased and practically identical if the mean is
greater than 20.
For data obeying the Poisson probability distribu-
tion, the standard deviation (SD) is given by SD = r ,
where r is the true mean. Because the true mean is usu-
ally estimated from an average of a number of individ-
ual measurements, the estimated standard deviation is
SD(est) = r (es).
Expressing standard deviation as a fraction or a per-
centage is often useful. The fractional standard devia-
tion is simply 1 r . The percent fractional standard

deviation (% SD) is the fractional standard deviation × 100.
For example, if 2500 counts are recorded in a picture ele-
ment or “pixel” in an image, the fractional standard devia-
1
tion of the measurement is 2500 = 0.2. The percent
fractional standard deviation is 2%. This equation can
also be expressed as:
%SD = 100
n Hendee WR: Medical radiation physics, 3rd ed. St. Louis,
where n is the number of counts observed.
Calculation of standard deviation is useful in deter-
mining the number of counts to obtain in the measure-
ment of a radioactive sample or in a scintigraphic image
for statistical certainty. As the number of counts
increases, the percent fractional or relative standard devi-
ation decreases and the ability to distinguish a true differ-
ence in the amount of radioactivity in two different
samples increases. Likewise, in imaging, as the number
of counts per pixel increases, the likeliness of the
observed differences in the image actually representing
Physics of Nuclear Medicine 33
true differences in the amount of activity between two
locations in the image increases.
SUGGESTED READING
Chandra R: Nuclear medicine physics: the basics, 6th ed.
Baltimore,Williams & Wilkins, 2004.
Mosby, 1992.
Johns HE, Cunningham JR: The physics of radiology, 4th ed.
Chicago,Thomas Books, 1983.
Powsner RA, Powsner ER: Essentials of nuclear medicine
physics, Malden, MA, Blackwell Science, 1998.
Cherry SR, Sorenson JA, Phelps ME: Physics in nuclear medi-
cine, 3rd ed. Philadelphia,WB Saunders, 2003.
Weber DA, Eckerman KF, Dillman LT, Ryman JC: MIRD: radionu-
clide data and decay schemes. New York, Society of Nuclear
Medicine, 1989.
 3
CHAPTER Radiation Detection
and Instrumentation

Radiation Detection documented, and the areas in which people work must
Ionization Chambers, Proportional Counters, and Geiger-Müller be monitored to ensure safety to both health care per-
Counters sonnel and patients. Radioactivity emanating from the
Basic Ionization Chambers patient must be detected to allow the temporal and spa-
Proportional Counters tial localization necessary to create scintigraphic images.
Geiger-Müller Counter The common denominator in all of the devices used in con-
Scintillation Detectors: Thallium-Activated Sodium Iodide Crystals temporary nuclear medicine practice for calibration of
Other Detection Devices administered dosages, area monitoring, and imaging is the
Gamma Ray Spectrometry and Pulse Height Analysis conversion of energy in the form of ionizing radiation into
Photopeak electrical energy. In modern imaging equipment,these elec-
Iodine Escape Peak tronic signals are often recorded and processed by dedicated
Compton Valley, Edge, and Plateau nuclear medicine computer systems. Nuclear medicine
Backscatter Peak imaging devices, including the gamma scintillation camera,
Lead Characteristic X-ray Peak can be thought of as specialized radiation detection devices,
Coincidence or Sum Peaks highly modified and adapted to record the temporal and
Compton Scatter in the Patient spatial localization of radioactivity in the patient.
Imaging Instrumentation
Rectilinear Scanners
Gamma Scintillation Cameras RADIATION DETECTION
The Patient as a Source of Photons
Collimators Ionization Chambers, Proportional
Gamma Ray Detection: The Sodium Iodide Crystal Counters, and Geiger-Müller Counters
Signal Processing and Event Localization One important approach to radiation detection is the use
Image Recording of an ionization chamber. The generic design concept is a
Characteristics of Modern Gamma Scintillation Cameras gas-filled chamber with positive and negative electrodes,
Gamma Camera Quality Control placed either at opposite sides of the chamber or in a con-
Field Uniformity centric cylinder geometry. A potential difference is cre-
Spatial Resolution and Linearity ated between the two electrodes, but no current flows in
Clinical Use of the Gamma Scintillation Camera the absence of exposure of the chamber to radiation. The
Window Setting interaction of ionizing radiation with the gas in the cham-
Computers in Nuclear Medicine ber creates positive and negative ions, which move to the
Creation of the Digital Image electrodes and produce an electrical current.
Data Analysis The basic concept of the ionization chamber is
Data Display and Formatting extremely versatile, allowing specialized devices to be
designed for specific applications. For example, the prob-
Detection of radioactivity is fundamental to the practice lem of detecting alpha and beta radiation is quite differ-
of nuclear medicine. The amount and type of radioactiv- ent from that of detecting gamma radiation because of
ity being administered to patients must be measured and differences in both their power to penetrate different

34

materials and their likelihood of interaction with matter.
In addition,the problem of surveying a wide area to deter-
mine the presence or absence of radiation is different
from the problem of accurately calibrating the millicuries
of activity to be administered to a patient. Three impor-
tant subtypes of ionization chambers with nuclear medi-
cine applications are the basic ionization chamber, the
proportional counter, and the Geiger-Müller counter.
Basic Ionization Chambers
The voltage difference between the electrodes in the
basic ionization chamber is calibrated to be just high
enough to “harvest” all of the ions from the sensitive vol-
ume of the chamber, but not high enough that the ions in
the chamber are accelerated to the point of creating addi-
tional secondary ionizations. As a result of this voltage
calibration strategy, the current produced in any single
event is very small and not measurable with any accuracy.
Rather, the ionization chamber is used to measure the
total current resulting from multiple events over a certain
integration time in a given radiation detection setting.
A number of devices routinely used in nuclear medi-
cine clinics operate on the principle of the ionization
chamber. Radiation survey meters such as the cutie-pie,
some pocket dosimeters, and radionuclide dose calibra-
tors are all examples of specialized basic ionization
chambers. The survey meters are typically calibrated to
provide units of exposure such as milliroentgens per
hour. Dose calibrators are set up to provide readings in
the units of radioactivity used in clinical practice. Many
laboratories now express these units in becquerels in
response to a mandate from the U.S. Food and Drug
Administration to use the International System as soon
as possible; other laboratories have retained the Ci, mCi,
and μCi convention. The amount of energy converted
to electrical current per unit of radioactivity is unique
for each radionuclide and radionuclide dose calibrators
must be calibrated for the radionuclide to be measured.
Proportional Counters
The main difference between a proportional counter and
the basic ionization chamber is greater applied voltage
between the electrodes in the former. The higher voltage
results in secondary ionizations in the sensitive volume of
the chamber. The term gas amplification describes this
phenomenon. Gas amplification can result in increased
ionization by a factor of 103–106. The resulting current
pulse is large enough to be measured individually and is
proportional to the energy originally deposited in the gas
chamber. Typically, an inert gas such as helium or argon is
used. The name of the device is based on the proportion-
ality of total ionization to the total energy of the ionizing
radiation. Proportional chambers do not have wide appli-
cability in clinical nuclear medicine. They are used in
research to detect alpha and beta particles.
Radiation Detection and Instrumentation 35
Geiger-Müller Counter
In the Geiger-Müller counter, the voltage is increased
even higher than in the proportional chamber applica-
tion. Because of the high voltage, the initial ionization
causes an “avalanche” of secondary ionizations, so that
the gas is essentially completely ionized. This mode of
operation of an ionization chamber allows detection of
individual events but not their energy (i.e., pulse count-
ing). Another important characteristic of the Geiger-
Müller counter is detector dead time. Because the gas in
the chamber is completely ionized, it takes a significant
amount of time to become ready for the next event.
Thus, Geiger-Müller counters are not useful in the pres-
ence of large amounts of radioactivity. They are good for
detecting low levels of activity and are widely used as
area survey meters and area monitors. They are valuable
in detecting radiation contamination.
Scintillation Detectors: Thallium-Activated
Sodium Iodide Crystals
The gas-filled ionization chambers described in the pre-
ceding section are not very sensitive to x-rays and gamma
rays because of the low likelihood of ionizing interac-
tions. The “stopping power”of gas is low. In current prac-
tice, thallium-activated sodium iodide crystals (NaI[Tl])
are used as the detector medium for single-photon imag-
ing systems. These crystals are optically transparent and
have sufficient stopping power for sensitive detection of
gamma rays (Table 3-1).
As noted earlier, an important common denominator
of many types of radiation detectors is the conversion
of the ionizing radiation energy to electrical energy.
Scintillation detector systems have an interesting conver-
sion process. Gamma rays or x-rays enter the sodium
iodide crystal and impart energy to valence electrons
during photoelectric and Compton interactions. The
imparted energy raises the electrons into the conduction
band of the crystal lattice. The energy difference between
the valence band and the conduction band is a few elec-
tron volts. As the electrons give up energy in the transi-
tion back from the conduction band to the valence band,
photons of light are emitted. The light photons have
a spectrum of energies; for sodium iodide crystals, the
spectrum peaks at a wavelength of 4150 Å, or approxi-
mately 3 eV. The energy conversion efficiency in the
NaI(Tl) crystal is approximately 13%. The remaining
energy is dissipated in the crystal in the form of molecu-
lar motion or heat. The scintillation decay time or length
of time for the scintillation event is approximately 1 μsec
(10–6 seconds).
Thallium-activated sodium iodide crystals have become
the preferred scintillation detector in many nuclear medi-
cine applications for a number of reasons. The crystals
are relatively inexpensive and allow great flexibility in
36 NUCLEAR MEDICINE: THE REQUISITES

Table 3-1 Half-Value Layers of Selected Radionuclides

Half-value layer (cm)

Radionuclide Energy (keV) Lead Water (soft tissue) NaI

Tc-99m 140 0.028 4.50 0.265

Tl-201

I-131
69

81
364
} Hg x-rays 0.0005

0.220
3.85

0.048
6.35
0.048

0.069
1.500

Collimator
Crystal

Positioning
pulses
Figure 3-1 Schematic of gamma scintillation
Position computer

X-ray or Gamma
camera. The diagram shows a photon reaching
gamma ray (X) the crystal through the collimator and undergoing
camera
photon oscilloscope photoelectric absorption. The photomultiplier
(Y) and tubes (PMTs) are optically coupled to the NaI(Tl)
computer crystal. The electrical outputs from the respective
memory photomultiplier tubes are further processed
through positioning circuitry to calculate (x, y)
coordinates and through addition circuitry to
calculate the Z pulse. The Z pulse passes through
the pulse height analyzer. If the event is accepted,
it is recorded spatially in the location determined
by the (x, y) positioning pulses.
PMTs Logic pulses

Addition Pulse
circuit height
analyzer

(z)
“Z” pulse

size and shape. The stopping power of the sodium
iodide crystals is good for the energy range used in clini-
cal nuclear medicine for single-photon applications (i.e.,
70–364 keV ) (see Table 3-1). The thallium impurities in
the sodium iodide crystal provide “activation centers” or
luminescence centers that offer easier pathways for the
return of the electrons from the conduction band of the
crystal to the valence bands of atoms requiring electrons
for electrical neutrality. Only a small amount of thallium
impurity (0.1–0.4 mole %) is required in the sodium
iodide crystal lattice to achieve the desired effect of mak-
ing the scintillation process more efficient. The conver-
sion efficiency of 13% is relatively high and the crystals
are internally transparent to the light photons so that
they reach the photocathodes. The disadvantages of
sodium iodide crystals are their fragility and their highly
hydroscopic nature, necessitating hermetically sealed
containers. In most applications, the crystal is sealed on
all sides by a thin aluminum canister except on the pho-
tomultiplier tube side, which is covered by a quartz win-
dow to allow the scintillation photons to escape and
reach the photomultiplier tubes.
The next step in the detection process is the interac-
tion of the light photons arising in the crystal with the
photocathode of a photomultiplier tube (Fig. 3-1). In the
typical sodium iodide detector system, whether it is
a simple probe or a gamma scintillation camera, the crys-
tal is optically coupled to the photomultiplier tube by
a light guide or light pipe to ensure the efficiency of light
collection. The light photons dislodge electrons from the

photocathode. These electrons are then accelerated by
a series of electrodes (dynodes) in the photomultiplier
tube.With each acceleration, the number of electrons is
increased. The electrons are collected at the anode or
collector of the photomultiplier tube. The multiplication
factor is on the order of 3–6 per dynode stage and up to
several million overall. The resulting voltage pulse from
the photomultiplier tube is then available for further pro-
cessing. This processing may take the form of amplifica-
tion followed by pulse analysis to determine either the
energy deposited in the crystal (pulse height analysis) or
the spatial location of the event (position analysis) in the
case of gamma scintillation cameras.
A key point to understand in the scintillation detec-
tion process is that proportionality is maintained at each
step. That is, the number of light photons given off in the
NaI(Tl) crystal is proportional to the energy deposited in
the crystal from the x-ray or gamma ray, the number of
electrons dislodged from the photocathode is propor-
tional to the number of light photons, and the electrical
output of the photomultiplier tube is proportional to the
number of electrons dislodged from the photocathode.
Thus, the height of the electrical pulse coming from the
photomultiplier tube is proportional to the energy of the
radiation absorbed in the crystal. This allows different
radionuclides with different energies to be distinguished
from one another by pulse height analysis. It also permits
a distinction between primary photons and photons that
have undergone Compton scatter events before detec-
tion. Compton-scattered photons are less energetic than
the primary photons and have lower pulse heights.
Recognizing Compton-scattered photons is critical in
imaging applications of scintillation detection because
only primary photons are desired to create the image.
Other Detection Devices
A host of other radiation detection devices are used in
nuclear medicine and radiology including photographic
film,which is used in personnel film badges,semiconduc-
tors, thermoluminescent and ultraviolet fluorescent
detection devices, and chemical detectors that are useful
for measuring cumulative radiation effects over a long
period. They are not discussed here.
GAMMA RAY SPECTROMETRY
AND PULSE HEIGHT ANALYSIS
The energies and relative abundance of the ionizing
radiations given off by each radionuclide are physical
constants. The proportionality between the energy of
a gamma ray and the output of the electrical pulse from
the photomultiplier tube provides a means for distinguish-
ing between gamma rays (or x-rays) of different energies.
Radiation Detection and Instrumentation 37
Figure 3-2 Spectrum for technetium-99m in air. The figure
illustrates the concept of full width at half maximum (FWHM). For
the particular detector system illustrated, the FWHM is 18 keV.
The energy resolution of the detector system for Tc-99m is 13%.
However, the spectrum of recorded energies is more com-
plex than would be predicted from the decay scheme
because of Compton and photoelectric interactions both
outside the NaI(Tl) scintillation detector and within the
crystal.Recognizing the consequences of these interactions
is important to the optimal use of counting and imaging
instrumentation.
By convention, the energy spectra from x-ray and
gamma ray detection are plotted with energy on the x-axis
and the relative number of events is plotted on the y-axis
(Fig. 3-2). The important relationships in gamma spectra
are illustrated here for technetium-99m (Tc-99m) because
it is the most commonly used radionuclide in clinical
practice.
Photopeak
In a perfect detection system and with the complete
absorption of the 140-keV gamma rays of Tc-99m in the
detector, a single line would be recorded on the energy
spectrum at exactly 140 keV.In practice,the 140-keV pho-
topeak is recorded as a bell-shaped curve centered at 140
keV (see Fig. 3-2). The Gaussian distribution of recorded
events stems from the statistical nature of the radiation
detection process. Each step in the conversion of ionizing
radiation to electrical current is subject to statistical fluc-
tuation. Light photons are given off in the scintillation
crystal with equal but random probability in all direc-
tions. Slightly different numbers of light photons impinge
on the photocathodes between different absorption
events. The number of electrons dislodged is also subject
to statistical fluctuation, as is the electron amplification at
each dynode stage in the photomultiplier tube.
The energy resolution of a detecting system can be
expressed by the spread in the photopeak. A frequently
38 NUCLEAR MEDICINE: THE REQUISITES
Figure 3-3 Energy spectrum for Tc-99m in air for a gamma
scintillation camera with the collimator in place. Note the iodine
escape peak at approximately 112 keV. The 180-degree
backscatter peak at 90 keV merges with the characteristic x-ray
peaks for lead (Pb). The Compton edge is at 50 keV.
used measure is full width at half maximum (FWHM),
which is defined as the energy range encompassed by
the bell-shaped curve halfway down from the apex of the
photopeak (see Fig. 3-2). A typical gamma scintillation
camera might have an FWHM equal to 14 keV for detect-
ing Tc-99m. Energy resolution of a detecting system can
also be expressed as a percentage of the photopeak
energy, and the detector would be said to have an energy
resolution of 10% (14/140).A narrower peak indicates
better energy resolution and a greater ability to distin-
guish gamma rays with energies close to each other. The
photofraction is the fraction of total counts in the entire
spectrum within the photopeak.
Iodine Escape Peak
Photoelectric interactions occurring close to the edge of
the sodium iodide crystal may result in the “escape” of
iodine K-characteristic x-rays from the crystal.When this
happens, the corresponding x-ray energy of approxi-
mately 28.5 keV is not deposited in the crystal and will
result in a small peak on the energy spectrum at 112 keV
(i.e., 140 − 28 keV) (Fig. 3-3). This peak, referred to as the
iodine escape peak, can be observed with a Tc-99m
source in air but is typically not observed in vivo because
of the relatively much larger contribution from
Compton-scattered photons from the patient in the
recorded energy spectrum (Fig. 3-4).
Compton Valley, Edge, and Plateau
Not every photon entering a NaI(Tl) crystal undergoes
photoelectric absorption. If a primary photon under-
goes a Compton scatter interaction in the crystal with sub-
sequent escape of the scattered photon, a smaller voltage
pulse will be detected than those composing the photo-
Figure 3-4 Compton scatter spectrum in soft tissue for single
scattering events. Note that Compton-scattered photons have
energy <140 keV but can be recorded above this level because of
the imperfect energy resolution of the gamma camera.
peak.If the 140-keV gamma rays from technetium are used
as the example, the maximum energy transferred to
a recoil electron in the crystal occurs at the largest scatter-
ing angle (180 degrees) and is 50 keV. This energy is
referred to as the Compton edge. The energy from 0–50
keV is called the Compton plateau or continuum and
corresponds to the energy deposited by photons that scat-
ter from 0–180 degrees before escaping the crystal (see
Fig.3-3). The portion of the energy spectrum between the
Compton edge and the photopeak is the Compton valley.
Some gamma rays undergo multiple Compton scatter
events before escaping from the detector crystal. These
events may be recorded in the region of the Compton val-
ley. The energy relationships obviously differ for each
radionuclide with differing photopeak energy.
Backscatter Peak
Another peak resulting from Compton scattering occurs
when primary gamma photons undergo 180-degree scat-
tering outside the detector and are then completely
absorbed. The scattering can take place either in front of
the detector or behind it if a gamma ray has initially
passed completely through the crystal without being
scattered or absorbed. From the previous section, it is
apparent that for Tc-99m the backscatter peak occurs at
90 keV (140 − 50 keV) (see Fig. 3-3).
Lead Characteristic X-ray Peak
In most nuclear medicine applications, scintillation
detectors are used in conjunction with lead collimators.
The 140-keV primary photons of technetium are ener-
getic enough to interact with the K shell electrons of
lead. The resulting K-characteristic x-rays are in the
range of 75–88 keV and are readily seen in the energy
spectrum (Fig. 3-5).

Figure 3-5 Energy spectrum from a gamma camera with the
Tc-99m activity in the patient. Note the loss of definition of the
lower limb of the Tc-99m photopeak. This spectrum can be
thought of as a sum of the spectra in Figs. 3-3 and 3-4. This
spectrum illustrates the difficulty of discriminating against
Compton-scattered photons using pulse height analysis.
Coincidence or Sum Peaks
The likelihood of two separate events taking place simul-
taneously in the sodium iodide crystal increases with the
amount of radiation present. If two events occur close
enough in time, the detector system may record them as
a single event. Two primary photons from Tc-99m that
are detected in coincidence will appear at 280 keV on
the energy spectrum. However, every combination of
events is possible. That is, a primary photon can be
detected in coincidence with a scattered photon of any
energy or a lead characteristic x-ray, and so forth. For
many detecting systems, the ability to discriminate or
resolve different discrete energies decreases with
increasing amounts of radiation exposure because the
likelihood of coincidence events increases with increas-
ing event rate.
Compton Scatter in the Patient
Degradation of clinical images is caused by Compton
scatter in the patient and the inability of imaging sys-
tems to completely discriminate primary from Compton-
scattered photons. For the gamma scintillation camera,
up to 35% (or even more) of recorded events come from
Compton-scattered photons. The energy spectrum for
Tc-99m photons undergoing one scattering event in the
patient ranges from 90 keV (i.e., 180 degree scattering
angle) to just under the energy of the primary photon,
140 keV (see Fig. 3-4). In Tc-99m spectra obtained with
radioactivity in the patient, the lower limb of the primary
photopeak merges into the events owing to Compton
scattering in the patient, which in turn merges with the
lead K-characteristic x-ray peak (see Fig. 3-5).
Radiation Detection and Instrumentation 39
IMAGING INSTRUMENTATION
The original instruments available for medical applica-
tions of radionuclides were handheld Geiger-Müller
devices and simple scintillation probe systems. These
systems did not allow spatial localization of radioactivity
emanating from the body but did provide a means of
crude overall counting. Early clinical applications in
nuclear medicine were aimed at calculating the percent-
age uptake of radioiodine in the thyroid gland with these
simple radiation detector systems.
Rectilinear Scanners
In the 1950s, probe systems were adapted into electro-
mechanical devices called rectilinear scanners. The geo-
metric field of view of the probe was focused or restricted
through the application of collimating devices and the
probes were mounted on mechanical transport systems to
systematically traverse back and forth over an organ of
interest. The original probe systems used calcium tungste-
nate crystals, which rapidly gave way to sodium iodide
crystals for the radiation detection step. By the 1960s, rec-
tilinear scanning systems were available with 3-, 5-, and 8-
inch diameter detectors.
Gamma Scintillation Cameras
The rectilinear scanner has been replaced by the gamma
scintillation camera invented by Hal Anger, also known as
the Anger camera. The gamma camera offers far more
flexibility than the rectilinear scanner and has been
developed into a sophisticated series of imaging devices
that permit dynamic and tomographic imaging, as well as
conventional static planar imaging. The major compo-
nents of the gamma scintillation camera are illustrated in
Fig. 3-1. Perhaps the easiest way to understand the way
gamma cameras work is to follow a photon through the
radiation detection and spatial localization process,
beginning with the origin of photons in the patient.
The Patient as a Source of Photons
Ideally, the flux of photons arriving at a radiation detec-
tor would be proportional to the number of photons
emitted in the respective part of the body being imaged.
This assumption would be valid only if the body part
were a point source of radiation in air, which is clearly
never the case in clinical practice. Additionally, a number
of factors cause distortion of the photon flux reaching
the gamma camera.
One may think of “good” photons as primary photons
arising in the organ of interest and emitted parallel to the
axis of the collimator field of view. These are the pho-
tons desired for creating the scintigraphic image. Good
40 NUCLEAR MEDICINE: THE REQUISITES
photons are reduced in number by absorption and scat-
ter, which decreases the information available for creat-
ing the image (Fig. 3-6). In the clinical applications of
nuclear medicine, many potentially useful photons are
absorbed or scattered before they reach the detector.
Unwanted primary photons can arise from back-
ground radioactivity in tissues in front of or behind the
structure of interest (see Fig. 3-6). These primary pho-
tons can travel directly to the detector and are then indis-
tinguishable from photons arising in the body part of
interest. They may be thought of as “bad” photons
because they reduce image contrast and may distort
quantitative data analysis. Background activity produced
by primary photons is hard to correct. One major advan-
tage of single-photon emission computed tomography
(SPECT) is the increase in image contrast resulting from
reduction in this kind of background activity superim-
posed on object activity.
Figure 3-6 Patient as a source of photons. A, Absorption and
scattering of primary photons in the body. These never reach the
detector. B, Background activity arising from in front of, behind,
and beside the organ of interest. C, Object and background
photons scattered toward the detector.
Another source of bad photons is primary photons
arising from the organ of interest, which travel “off axis”
toward the detector. Radiation is given off isotropically
(i.e., with equal probability in all directions) and only
a small fraction of the total emitted photons are useful for
forming the image. A principal function of collimators is
to absorb off-axis photons (Fig. 3-7).
Compton scatter is a third source of bad photons (see
Figs. 3-6 and 3-7). Photons originating in or adjacent to
the organ of interest can scatter and subsequently travel
toward the detector. Photons that undergo Compton
scattering in the patient lose some of their energy and
can be partially discriminated against by using pulse
height analysis. However, this ability is far from perfect.
For example, a 140-keV photon scattered through
a 30-degree angle retains an energy of 135 keV. This energy
would be accepted in a typical 20% energy window used
for clinical imaging with Tc-99m.
Figure 3-7 Interaction of photons arising in the patient with
detector and parallel-hole collimator. The collimator provides
directional discrimination for primary and scattered photons. It
does not eliminate either background or scattered photons that
travel toward the detector within the geometric acceptance field
of view of the collimator.“Good”photons are primary
(unscattered) photons that originate in the object and travel
parallel to the axis of the collimator field of view.All other photons
(i.e., background, scattered, off-axis) are undesirable in the image.
 Radiation Detection and Instrumentation 41

Collimators ventionally to refer to collimators designed for photons

The collimator is the first part of the gamma scintillation of the energy of Tc-99m (140 keV) or less. Medium-
camera potentially encountered by the photon after it energy collimators are designed for radionuclides with
leaves the patient (see Fig. 3-1). The purpose of the colli- gamma emissions less than 400 keV, such as gallium-67
mator is to define the geometric field of view of the crys- (Ga-67) (multiple photopeaks at 93, 185, 300, and 395
tal and specifically to define the desired direction of keV ). The septa—the lead wall between collimator
travel of gamma rays allowed to reach the crystal (see holes—is relatively thicker in these collimators than in
Fig. 3-7). Because x-rays and gamma rays in the energy low-energy collimators. Although the principal gamma
range used in nuclear medicine cannot be focused using ray from iodine-131 (I-131) (364 keV) falls in this energy
a lens, the only way to determine the directionality of the range, the presence of several higher energy photons
photons is by absorptive collimation,that is,by absorbing (>600 keV) in the decay scheme results in significant
all of the photons that are not traveling in the proper degradation and septal penetration artifacts. This is
direction. Absorptive collimation is very inefficient caused by photons from one hole crossing the septa into
because only a small fraction (about 1 in 10,000) of the
emitted photons will be traveling in the proper direc-
tion. The collimator discriminates against unwanted
photons only on the basis of their direction of travel. The
collimator does not distinguish between primary and
scattered photons or among photons of different ener-
gies. The pulse height analyzer is used to discriminate
against scattered photons and other photons of unwanted
energy that reach the detector. Collimators for gamma
scintillation cameras are available in four basic types:
pinhole, parallel hole, converging, and diverging.
Pinhole Collimators
The geometric behavior of a pinhole collimator is similar
to that of a pinhole camera (Fig. 3-8). Beyond the focal
point, the field of view increases with distance and the
image is inverted. The principal use of the pinhole colli-
mator is for imaging small objects such as the thyroid and
bones of the hand and feet. In this application, it offers
the advantage of image magnification when the aperture-
object distance is shorter than the distance from the
detector to the aperture. The geometric magnification
allows the resolution of objects smaller than the intrinsic
resolution of the gamma camera, which is particularly
valuable for thyroid imaging, in which lesions as small as
3–5 mm may be resolved. The pinhole collimator also
offers flexibility in patient positioning and is useful in
obtaining oblique views of the thyroid. Pinhole collima-
tors have also been used to magnify small structures in
pediatric patients (e.g., the hip joint).
The major disadvantage of the pinhole collimator is
poor count rate sensitivity. The usual pinhole aperture
diameter is 3–6 mm. Any increase in this size to increase
count rate results in a corresponding decrease in spatial
resolution.
Parallel-Hole Collimators
The parallel-hole collimator is the workhorse collimator Figure 3-8 Pinhole collimator. The image is inverted. The
in daily nuclear medicine practice (Fig. 3-9). Typically, image is magnified if the distance from the aperture to the object
such collimators consist of lead foil with thousands of is smaller than the distance from the aperture to the gamma
camera crystal. The object is minified if its distance from the
parallel channels or holes uniformly distributed. A num- aperture exceeds the aperture-to-crystal distance. Spatial
ber of terms are used to further characterize parallel-hole resolution and count rate sensitivity are inversely affected by
collimators. The term low-energy collimator is used con- aperture diameter.
42 NUCLEAR MEDICINE: THE REQUISITES
Figure 3-9 High resolution collimator. The upper collimator
with longer septa is designed to achieve higher resolution. Septal
thickness and energy rating are the same.
the neighboring hole when I-131 is used with medium-
energy collimators. Thus, high-energy collimators have
been designed for I-131. These collimators have thicker
septa than low- and medium-energy collimators. Special
collimators made of tungsten or other high-Z material
have been fabricated for imaging the 511-keV photons of
positron emitters.
Among low-energy collimators, designs are opti-
mized for either sensitivity or resolution. For a given
hole size and septal thickness, a thicker collimator (i.e.,
longer holes) results in higher spatial resolution and
lower sensitivity. Longer holes of equal diameter have a
smaller acceptance angle, resulting in a loss of count
rate sensitivity but an improvement in geometric or spa-
tial resolution (see Fig. 3-9).
One useful way to think about parallel-hole collimators
is to look in detail at the geometry and characteristics of an
individual hole. Conceptualized in this way, the field of
view of each hole becomes larger with increasing distance
from the collimator face (Fig. 3-10). Fields of view from
adjacent holes begin to overlap and geometric resolution
is degraded.For this reason,the organ of interest should be
positioned as close to the collimator surface as possible
because that is where the resolution is best. Unlike pin-
hole collimators, image size is not affected by collimator-
to-source distance with parallel-hole collimators.
The physics of collimator design and response is com-
plex. In essence, a trade-off occurs between spatial reso-
lution and count rate sensitivity. There is also a need to
avoid excessive septal penetration. The need to mini-
Figure 3-10 Detail geometry of a single-hole collimator. The
field of view increases with distance from the collimator face.
mize septal penetration defines the choice of septal
thickness for a given energy (see Table 3-1). The rule of
thumb in collimator design is that less than 10% of
recorded events should be from septal penetration. The
off-axis photons reaching the scintillation crystal by sep-
tal penetration degrade spatial resolution. However,
thicker septa reduce sensitivity and observed count rate.
It is also important to note that the sensitivity of a parallel-
hole collimator does not change with distance.Although
the sensitivity of a single hole will decrease with dis-
tance,more holes will be irradiated yielding no net change
in sensitivity with distance.
Converging-Hole and Diverging-Hole Collimators
Converging-hole collimators are used to magnify the
image geometrically (Fig. 3-11). They are applied mostly
in pediatric nuclear medicine, where they have partially
replaced pinhole collimators for this purpose. Fan beam
and cone beam collimators used in SPECT are special
adaptations of converging collimators that optimize use
of the detector surface area. In SPECT, the use of these
collimators leads to an increase in sensitivity without
a subsequent loss in spatial resolution.
Diverging collimators were popular before large-field-
of-view cameras became available (Fig. 3-12). They per-
mit a larger area of the body to be imaged than is possible
with a parallel-hole collimator. For example, a lung scan
of a large patient is not feasible with a gamma camera
having a standard 10-inch-diameter field of view but is
readily accomplished with a diverging-hole collimator.
The main drawback of converging and diverging colli-
mators is distortion of the image, which occurs because
each portion of the organ of interest is magnified or

Figure 3-11 Converging-hole collimator. Objects are magnified.
minified to a different extent, depending on the distance
between the respective location and the collimator.
In addition to the primary collimator designs, a num-
ber of specialty use collimators have been described.
Parallel slant-hole collimators have found application in
nuclear cardiology. Some nuclear medicine physicians
favor a 30-degree caudal angulation for separating the left
atrium from the left ventricle in radionuclide ventriculog-
raphy. Rotating slant-hole collimators and multiple-pin-
hole collimators have been used for limited angle
emission computed tomography.
Gamma Ray Detection: The Sodium Iodide Crystal
Modern gamma scintillation cameras use thallium-
activated sodium iodide crystals as the radiation detector
(see Fig. 3-1). The desired event in the camera crystal is
the complete photoelectric absorption of a primary pho-
ton that reached the crystal by traveling parallel to the
geometric axis of the collimator field of view from its ori-
gin in the organ of interest in the patient. The likelihood
of a photoelectric interaction and complete energy
absorption in the sodium iodide crystal is greater at low
energies and decreases at higher energies as Compton
scatter becomes more likely (see Table 3-1).
Radiation Detection and Instrumentation 43
Figure 3-12 Diverging-hole collimator. Objects are minified.
As discussed in the section on radiation detection, the
gamma ray energy is converted to light energy in the
crystal. For every 140-keV technetium photon com-
pletely absorbed, approximately 4200 light photons are
emitted with an average energy of 3 eV. One of the limita-
tions of lower energy gamma rays, including those from
Tc-99m, is the limited number of light photons available
for subsequent event localization.Higher energy photons
potentially provide more light photons and better statisti-
cal certainty for event localization. However, this is coun-
terbalanced by the greater likelihood of an initial
Compton scatter event in the crystal before a terminal
photoelectric interaction. When multiple scattering
events occur in the crystal before complete energy
absorption, spatial resolution is reduced.
Signal Processing and Event Localization
The breakthrough concept in the design of the gamma
scintillation camera is the use of an array of photomulti-
plier tubes behind the crystal for event localization. In
the first commercial gamma camera, a 10-inch diameter
sodium iodide crystal was optically coupled to a hexago-
nal array of 19 3-inch diameter photomultiplier tubes
(Fig. 3-13).
44 NUCLEAR MEDICINE: THE REQUISITES
Figure 3-13 Circular (A) and hexagonal (B) photomultiplier
tubes. The tubes are arrayed in a hexagonal configuration so that
the distance from each tube to all of its nearest neighbors is
identical. The switch from round to hexagonal tubes allows more
complete coverage of the gamma camera crystal.
For each event, two kinds of signal processing are per-
formed. First, the output from all of the photomultiplier
tubes is summed for the purpose of pulse height analysis.
This summed pulse is typically referred to as the Z pulse.
It is used to determine whether the detected event
is within the desired energy range and should be
accepted into the formation of the image. If it is of lower
or higher energy, it is discriminated against and rejected
(see Fig. 3-1).
Simultaneously, the output of each photomultiplier
tube is looked at in a different way. Each tube may be
thought of as having x and y coordinates in a Cartesian
plane, with the center of the central photomultiplier
tube being the origin. Each photomultiplier tube then
can be thought of as contributing either a positive or
a negative value for x and y positioning. The photomulti-
plier tubes closest to the event collect the greatest num-
ber of light photons, with lesser contributions from more
remote tubes. The logic circuitry of the camera is used to
compute the most likely coordinates of the event loca-
tion in the crystal by adding together all of the x and y
pulses from the 19 photomultiplier tubes (see Fig. 3-1).
Image Recording
If the Z pulse indicates that a primary photon has been
absorbed, an unblanking signal is sent to the image-
recording device. On original cameras, the recording sys-
tem was an oscilloscope with a Polaroid camera or
35-mm camera attachment. The x and y positioning sig-
nals provided the deflection coordinates for the cathode
ray tube (CRT) and the event was recorded on film as
a single flash of light from the screen. In modern cam-
eras, the signals from the photomultiplier tubes are indi-
vidually digitized by analog-to-digital converters (ADCs).
The Z signal and the position signals are determined by
a computer.If the Z signal is within the energy acceptance
window (e.g., 133–147 keV for Tc-99m), the pixel in the
computer matrix corresponding to the estimated position
of the event is incremented. A typical image is created by
recording 100,000–1,000,000 individual events.
Characteristics of Modern Gamma
Scintillation Cameras
The original commercial gamma cameras had 10- to
12-inch-diameter crystals with a thickness of 0.5 inch.
These cameras were designed in an era when I-131 (364
keV ) was the most important radionuclide. In the ensu-
ing 25 years, crystal sizes and shapes have changed. Large
field-of-view cameras with 30–50 cm rectangular fields of
view have become the standard.
A series of changes in the original gamma camera
design has been aimed at improving spatial resolution.
The crystal thickness can vary in modern cameras from
0.25–1 inch. The 0.25-inch crystal thickness is more
suited to studies with lower energy radionuclides, such
as Tc-99m and thallium-201. For Tc-99m with a 140-keV
principal photon energy, the loss in sensitivity between
0.5- and 0.25-inch thickness is only 6% (see Table 3-1),
whereas the spatial resolution is improved by 20%. For
Tl-201, there is virtually no loss of sensitivity with the
same 20% improvement in spatial resolution. However,
for studies using gallium (93, 185, 300, 394 keV), indium-
111 (172, 247 keV), or I-131 (364 keV), a 3⁄8-inch thick
crystal is most commonly used.
SPECT imaging of fluoride-18 rekindled interest in
thicker crystals. The detection efficiency for 511-keV
photons increases from ~12% for 3⁄8-inch NaI(Tl) crystals
to ~18% for 0.5-inch crystals. Crystal thicknesses of up to
1 inch are used in some cases.
The number of photomultiplier tubes used in gamma
cameras has increased. The first step was to reduce tube
diameter from 3 to 2 inches, which permitted use of 37
photomultiplier tubes for a standard field-of-view cam-
era. Large field-of-view cameras are available with 55, 61,
75, and even 91 tubes.Another advance in photomulti-
plier tubes is the hexagonal photocathode, which allows
the tubes to cover the crystal completely without leaving
gaps between them (see Fig. 3-13). Light pipes have been
replaced with direct coupling of the photomultiplier
tubes to the crystal. The collection of more light photons
reduces the statistical uncertainty in the (x, y) event
localization logic circuitry.
An area that has received major attention over the
years is field uniformity. The basic problem is that each
photomultiplier tube behaves slightly differently and
may drift in its performance over time. Field uniformity
was not a major problem before SPECT but is now criti-
cal to prevent artifacts in SPECT images.
In addition to slight differences in photomultiplier
tube response, subtle differences occur in the crystal
itself and in the efficiency of the optical coupling of the
photomultiplier tubes with the crystal. The collection
efficiency is also very dependent on whether the initial

Figure 3-14 Slightly offset spectra from two different
photomultiplier tubes in the gamma camera crystal and the
combined spectrum. Especially in older gamma scintillation
cameras, a wide energy acceptance window was necessary to
encompass the variations in response across the crystal.
interaction occurred right above a photomultiplier tube
or between tubes. For this reason, the energy spectrum
that is collected from any one photomultiplier tube is dif-
ferent from all the other tubes (Fig. 3-14). The observed
energy spectrum from the overall camera is made up of
a sum of the slightly different spectra from each tube.
This spectrum could be dramatically demonstrated in
some cameras by setting an asymmetrical pulse height
analyzer window over a photopeak to accentuate the dif-
ferences in tube performance.
Although vendors have tried a number of approaches
to match the performance characteristics of the photo-
multiplier tubes, problems persist. The current approach
is to use computer correction of the response across the
crystal. In effect, after the camera system is manufactured
and tuned as well as possible, its actual performance rela-
tive to a known radioactive source energy and its imag-
ing geometry are empirically mapped and correction
factors are established for each small area of the detector.
The ZLC system introduced by Siemens a number of
years ago is illustrative of attempts to correct for spatial
variation in energy response and for small nonlinearities.
The Z signals are corrected by empirically measuring
a 128 × 128 energy response matrix. Each Z signal is then
corrected by a factor, ΔZ, obtained for its respective pixel
location in the matrix before the pulse reaches the pulse
height analyzer. The corrected pulse (Z + ΔZ) is then ana-
lyzed. Each event is energy corrected on the fly during
image acquisition.
For linearity correction, a rectilinear grid is imaged, and
a 4k × 4k lookup table of correction factors for spatial local-
ization is created. Each event is positioned in the image
based on ΔX and ΔY correction factors corresponding to
the observed location of the event.
Radiation Detection and Instrumentation 45
Figure 3-15 Four-quadrant bar phantom images obtained with
the gamma camera stationary and during rotation. Note the
degradation in bar phantom resolution in this early generation
rotating SPECT system.
More recently, several strategies for automatic and
active photomultiplier tube adjustment have been intro-
duced. In one system, light-emitting diodes of known out-
put are used to measure and fine-tune photomultiplier
tube response as often as 10 times per second. This
approach is advantageous for applications in which the
camera head is rotating, because photomultiplier tube
performance can be affected by changes in alignment to
the earth’s magnetic field (Fig. 3-15). Each gamma camera
vendor has taken a different approach to the energy
response and spatial localization problems. The unifying
theme is increasing sophistication in making corrections
event by event.
In the best contemporary cameras, the recording of
each event is corrected separately for location and
energy. This kind of event-by-event correction permits
the use of asymmetrical windows. The advantage of an
asymmetrical window offset to the high side of the pho-
topeak is reduction in scattered photons accepted in the
image. However, unless energy correction is performed
properly,the response across the image will vary depend-
ing on photomultiplier tube response. Events in areas of
lower output tubes will be underrepresented in the
image, whereas events in areas with higher output tubes
will be overrepresented (Fig. 3-16). Further advances
from commercial vendors have led to automatic tuning
systems and online adjustment systems for photomulti-
plier and overall system response.
The past 15 years has seen an explosion in the num-
ber and kinds of gamma cameras on the commercial
market. Mobile cameras, whole body imaging systems,
and cameras adapted to special nuclear cardiology appli-
cations are available, as are camera systems with multi-
ple detector heads for SPECT and whole body imaging.
46 NUCLEAR MEDICINE: THE REQUISITES
Figure 3-16 Effect of different photopeak window settings.
All images are from the same patient. A–D, The energy spectrum
from the patient. The location of the energy window is indicated
by the black rectangle superimposed over the spectral lines. E–H,
The resulting liver image. E, A symmetrical window centered at
the proper photopeak. F–G, The energy window is offset to the
high side. H, The window is offset to the low side. Note the loss of
homogeneity in the liver in F and G with a geometric pattern of
hot and cold areas owing to the location pattern of the
photomultiplier tubes. Scatter is decreased, as indicated by the
lower counts coming from the heart region, but the images are
grossly misleading. In H, the image is degraded by excessive scatter
and loss of spatial resolution. Note the blurring of the liver margin.
Gamma Camera Quality Control
Gamma scintillation cameras are complex devices
with physical, mechanical, and electronic components.
Malfunction or breakage of any of these components
can be catastrophic to system performance and may not
be recognized from a review of clinical images. For
these reasons, a number of comprehensive and sophisti-
cated procedures have been developed over the years
to ensure adequate camera performance. The ones
used most often in routine clinical practice are summa-
rized in Box 3-1. In addition to these, the National
Electrical Manufacturers Association (NEMA) has devel-
oped a comprehensive set of tests to measure camera
performance.
Field Uniformity
One fundamental parameter that requires daily assess-
ment is the uniformity of response of the gamma camera
across its entire field of view (Fig. 3-17). A source of
radioactivity of appropriate energy is used to test the
camera response. Measurements made with the collima-
tor in place are referred to as extrinsic, and those made
without the collimator are referred to as intrinsic.
The specific method for assessing field uniformity
varies. For example, a uniform disk or flood source in
a phantom can be used to measure extrinsic field unifor-
mity. With this approach, the radioactive source is placed
at or on the surface of the gamma camera collimator. To
measure intrinsic field uniformity, a point source of
radioactivity is positioned at the center of the crystal at
a distance from the uncollimated crystal face. The rule of
thumb is that the source should be at a distance at least
equal to five times the size of the field of view to acquire
a uniform image. For example, if the camera has a 40-cm
field of view,the point source should be placed at least 200
cm away. In the case of fixed dual-detector SPECT sys-
tems, it is impossible to get the source the necessary dis-
tance from the camera, and thus the acquired image has
higher counts in the center than on the edges.In this case,
a correction can be applied to correct for this geometric
nonuniformity so that instrumentational nonuniformities
can still be evaluated. Typically, 1000k–5000k counts are
obtained to evaluate field uniformity for planar imaging.
For extrinsic field uniformity testing, most laboratories
use either a phantom filled with a uniform solution of
Tc-99m or a permanent disk source of uniformly distrib-
uted cobalt-57 (Co-57; T1⁄ 2 270 days,122 keV ). The standard
practice is to obtain a flood image with each camera every
day before it is used for clinical studies. In laboratories
where obtaining the daily flood image with the collimator
in place is more practical,obtaining an intrinsic flood image
weekly is still useful,and vice versa for laboratories that rou-
tinely acquire flood images without the collimator in place.
 Radiation Detection and Instrumentation 47

BOX 3-1 Gamma Camera Quality Control Summary

Parameter Comment

DAILY
Uniformity Flood field; intrinsic (without collimator) or extrinsic (with collimator)
Window setting Confirm energy window setting relative to photopeak for each radionuclide used with each
patient

WEEKLY
Spatial resolution Requires a “resolution”phantom (parallel line equal spacing, four-quadrant bar, orthogonal
hole) and standardized protocol
Linearity check Qualitative assessment of bar pattern linearity

PERIODIC (BIANNUALLY OR WHEN A PROBLEM IS SUSPECTED)

Collimator performance
Energy registration
Count rate performance
and count rate linearity
Energy resolution
Sensitivity
High count flood with each collimator
For cameras with capability of imaging multiple energy windows simultaneously
More important in cameras with “count skimming”or “count addition”correction circuitry
Easiest in cameras with built in multiple-channel analyzers
Count rate performance per unit of activity

Figure 3-17 Flood source and four-quadrant bar phantom.

Images of the flood source (A) and bar phantom (B) from a well-
tuned gamma scintillation camera with the collimator off. The
flood image shows slight mottling but no focal or localized areas of
increased or decreased activity within the center of the field of
view. The slightly increased activity along the rim is a common
characteristic of gamma cameras seen on intrinsic flood images.
The smallest bars are partially discernible on the bar phantom
image. They have a spacing of 3 mm. The bar images show good Figure 3-18 Abnormal flood. Image from a camera with a
linearity. nonfunctioning central photomultiplier tube and a crystal defect.

The image obtained in the field uniformity examina-
tion should be carefully inspected. A well-tuned camera
with proper photomultiplier tube and correction cir-
cuitry performance should provide a flood image with
a highly uniform appearance. Some minor mottling with
slightly increased intensity in regions corresponding to pho-
tomultiplier tubes is acceptable (Fig. 3-17). Photomultiplier
tube drift or even the failure of a photomultiplier tube can
be recognized as an area of decreased activity (Fig. 3-18).
Cracked crystals are readily identified and even damage to
a collimator can be detected. The soft lead in collimators
is often protected by a covering but can still be subject to
denting, causing bending and distortion of the septa.
Spatial Resolution and Linearity
Bar phantoms are routinely used to evaluate image reso-
lution and linearity in the clinic setting. With modern
gamma cameras, a weekly assessment is sufficient. The
phantoms are constructed of parallel lead strips encased
in a plastic holder. Resolution is defined by the ability to
discriminate between two distinct points. For routine
clinical gamma camera quality control, visual assessment
48 NUCLEAR MEDICINE: THE REQUISITES
is adequate. The subjective spatial resolution is expressed
in terms of the smallest bar pattern visible on the image.
In a properly functioning camera,all groups of bars in the
bar phantom pattern should appear straight and parallel
(see Fig. 3-17). Some distortion is typically seen at the
edge of the field of view.
A four-quadrant bar phantom is most commonly used.In
this phantom, the lead strips are thinner and spaced closer
in sequential quadrants (see Fig.3-17). The phantom is cho-
sen so that the quadrant with the narrowest lead strips
appears slightly blurred. The phantom is positioned on the
collimator face with the center of the four-quadrant pattern
corresponding to the center of the camera. A uniform Tc-
99m flood source is then placed on the bar phantom. Four
images are obtained at sequential 90-degree rotations
between positions. Care must be taken not to position the
bars on the collimator in such a way that an interference or
moiré pattern occurs (Fig.3-19).
An alternative is the parallel-line equal-spacing bar
phantom.When this device is used, two images are nec-
essary. Because signals from the photomultiplier tubes
are processed through two essentially independent
positioning circuits (x and y), degradations can occur
in a selective direction. Another alternative is the
orthogonal-hole test pattern (see Fig. 3-19). It is
designed so that only a single image is required.
Regardless of the phantom chosen, when trouble is sus-
pected, the procedure should be repeated with and
without the collimator.
The spatial resolution of gamma cameras is
expressed quantitatively as the full width at half maxi-
mum (FWHM) of a line spread function. A line spread
function is obtained by first imaging a narrow line
Figure 3-19 Moiré patterns. Note the abnormal pattern in the
three triangles on the left in an image of a “hot spot”phantom. The
distortion is especially marked in the lower left triangle.
source of radioactivity on the collimator (extrinsic) or
crystal face (intrinsic), followed by determining a count
profile or histogram perpendicularly across it. This his-
togram is called the line spread function. In an imag-
ing system with perfect spatial resolution, the line
spread function would have a single spike correspon-
ding to the radioactive line source. In practice, a bell-
shaped curve is seen.
FWHM is simply the distance encompassed by the
curve halfway down from its peak. This measurement is
the same as previously discussed for describing energy
resolution. By analogy, a narrower peak indicates better
spatial resolution and therewith the ability to resolve
objects close to each other. In modern gamma cameras,
intrinsic resolution (collimator off) approaches 3-mm
FWHM or less.An estimate of FWHM can be made using
the four-quadrant bar phantom by determining the small-
est discernible bars and then multiplying the size of the
bars by a factor of 1.7.
Clinical Use of the Gamma
Scintillation Camera
Applying the gamma scintillation camera to clinical pro-
cedures requires the development of imaging protocols
that define the diagnostic purpose, the radiopharmaceu-
tical to be used, patient preparation, and the imaging
sequence. These issues are discussed in the organ system
chapters for the major scintigraphic studies. The proto-
cols include selection of collimator, timing of image
acquisition after radiopharmaceutical administration,
time per image or number of counts to be recorded, and
actual images or views to be obtained.
Window Setting
A quality control issue sometimes overlooked in the clini-
cal application of gamma cameras is the setting of the
energy window. The most common approach is to use
a symmetrical window centered at the energy peak of the
radionuclide label being used in the imaging procedure.
For Tc-99m, the most common recommendation is to use
a 20% window centered at 140 keV. The acceptance
range for this window is 126–154 keV. In gamma cameras
with energy correction circuitry, setting an asymmetrical
window to reduce Compton scatter may be possible.
Using a narrower window of 10–15% for higher resolu-
tion imaging may also be desirable. These approaches
should be undertaken with caution for older gamma cam-
eras because of the problem of nonuniform response
across the crystal, which is discussed in some detail in
a previous section.
The most conservative approach is to confirm the
window setting for each radionuclide used during
the course of a day and then to reconfirm the window

setting before imaging each patient. Setting the energy
window (“peaking in the camera,”“setting the peak”)
should be done with a radioactive source in air and not
by using radioactivity in the patient. The spectrum from
the patient includes scatter that can shift the perceived
location of the photopeak.
False positive and false negative interpretations may
occur because of artifacts and loss of resolution, respec-
tively, with incorrect window settings. Occasionally, the
window is inadvertently left at the setting for a Co-57
flood source (122 keV). Figure 3-20 illustrates the degra-
dation of image quality in a Tc-99m diphosphonate bone
study resulting from this error.
Another practical problem of window setting occurs
in cameras that image multiple photopeaks simultane-
ously. Care must be taken to ensure that the image data
from the different photopeaks are correctly registered
together on the clinical image. Figure 3-21 illustrates
incorrect and correct multipeak registration for a Ga-67
flood image.
Figure 3-20 Incorrect photopeak setting. A, Image obtained
with a 20% window set at 122 keV, the energy of the cobalt-57
flood source. The image quality is dramatically improved in B,
which was obtained at the correct window setting for technetium-
99m.
Figure 3-21 Spatial misregistration. Gallium-67 flood images
obtained using multiple photopeaks. A, Artifacts in the flood
image caused by spatial misregistration. B, The properly registered
image shows good uniformity.
Radiation Detection and Instrumentation 49
COMPUTERS IN NUCLEAR MEDICINE
Computers in nuclear medicine were first used clinically
with the development of gated blood pool imaging in the
mid-1970s. Subsequently, the computer has become a pri-
mary image acquisition and processing device and it is
used for image management and formatting, in addition
to its integral role for dynamic studies and SPECT.
Currently, all state-of-the-art gamma cameras have com-
puters as fundamental components of the system.
Creation of the Digital Image
The x and y pulses generated in the gamma scintillation
camera logic circuitry define event location. In older
cameras, these pulses are in analog form and must be
converted to digital form for computer processing. To
accomplish this, an analog-to-digital converter is inter-
posed between the gamma camera and the computer.
Some modern cameras convert the (x, y) signals to digital
form within the camera’s own electronic circuitry. The
Z pulse is used in computer data acquisition to indicate
that a particular event should be accepted for storage.
Two fundamentally different modes have been used to
acquire and store digitized data: list (serial) mode and
frame (histogram) mode. In the list mode approach, each
pair of digitized (x, y) position signals is stored separately
and sequentially in computer memory. The “list” is simply
a line of data flowing into computer memory. If time infor-
mation is desired, additional time markers are inserted
into the list. Physiological signals such as the R wave on
the electrocardiogram can also be recorded (Fig. 3-22).
The list mode approach offers great flexibility. For
example, data from each cardiac cycle can be analyzed
separately. If a particular beat was caused by an arrhyth-
mia, the data from that beat can be excluded from the
desired data from normal sinus beats.Alternatively, data
from beats caused by particular types of arrhythmias can
be analyzed separately. The major disadvantage of list
mode is that it requires a large amount of computer
memory to store the study data.It also requires additional
time to process data into an image format after acquisi-
tion is complete.
In the alternative histogram or frame mode of data
acquisition the digitized (x, y) pairs are used to locate the
picture element to which they belong. The image may be
thought of as a grid or matrix superimposed on the ana-
log data (Fig. 3-23). The x and y numbers determine
which matrix element encloses the location of the origi-
nal event.At the end of data collection,rather than having
discrete information on each event, each matrix location
has a number corresponding to the total events accumu-
lated throughout the imaging period.
50 NUCLEAR MEDICINE: THE REQUISITES
Figure 3-22 List mode data acquisition.
Figure 3-23 Digital image. Analog image (A) has 6 × 6 matrix
superimposed (B). The number of events (dots) in each pixel is
recorded to create the digital matrix (C).
Frame mode is much more sparing of computer mem-
ory. It has the further advantage that the data are immedi-
ately ready for display or analysis, without postprocessing
or formatting. Physiological signals can still be used to
control data collection as in multigated cardiac studies.
However, once recorded, data from arrhythmic beats can-
not be excluded. Dynamic studies can be acquired by
using multiple frames as a function of time.
A larger matrix results in better potential spatial reso-
lution but also requires longer time to achieve adequate
counting statistics in each pixel. Most studies in current
practice are obtained in a 128 × 128 matrix, although
64 × 64 and 218 × 218 matrixes are also used.
Data Analysis
Computer recording of image data greatly facilitates
quantitative analysis. Specific types of analyses are dis-
cussed in the respective organ system chapters. A recur-
ring requirement in data analysis is the definition of
a“region of interest.” These regions can be defined by the
computer operator or through the use of automated
region of interest definition programs. The latter are
often used to define the area of the left ventricle of the
heart in calculating ejection fractions.
The computer can make various calculations on the
pixels in regions of interest. In most applications, the
total count within the region is of greatest value. This
kind of data analysis allows calculation of quantitative
parameters such as the left ventricular ejection fraction
or the percentage of the total glomerular filtration rate
attributable to the left versus the right kidney.
Data Display and Formatting
Clinics with contemporary computer systems frequently
use them to archive image data and to control image-
formatting devices such as laser printers. There are
advantages of using the computer for this purpose rather
than using analog imaging or recording directly from the
gamma camera cathode ray tube. Images may be win-
dowed and centered to provide the optimum gray scale
after the fact. Also, the same image data may be viewed

with and without secondary image processing, including
background subtraction or contrast enhancement. It is
also possible to view correlative images for other modali-
ties on the system. For example,a SPECT brain scan can be
compared to an MR image acquired on the same patient.
The computer is invaluable for looking at dynamic data.
This capability is most important for viewing the beating
heart in nuclear cardiology. It also has value in perform-
ing time lapse photography for other applications, such
as localizing the site of bleeding in gastrointestinal bleed-
ing detection studies or assessing biliary dynamics dur-
ing hepatobiliary imaging studies.
SUGGESTED READING
Chandra R: Nuclear medicine physics: the basics, 6th ed.
Baltimore,Williams & Wilkins, 2004.
Radiation Detection and Instrumentation 51
Hendee WR:Medical radiation physics, 3rd ed.St.Louis,Mosby,
1992.
Hutton BF, Barnden LR, Fulton RR: Nuclear medicine comput-
ers: applications. In Nuclear medicine in clinical diagnosis
and treatment, 3rd ed. Ell PJ, Gambhir SS, Eds. New York,
Churchill Livingstone. 2004, pp 1793-1814.
Johns HE, Cunningham JR: The physics of radiology, 4th ed.
Chicago,Thomas Books, 1983.
Powsner RA, Powsner ER: Essentials of nuclear medicine
physics, Malden, MA, Blackwell Science, 1998.
Cherry SR, Sorenson JA, Phelps ME: Physics in nuclear medi-
cine, 3rd ed. Philadelphia,WB Saunders, 2003.
Weber DA, Eckerman KF, Dillman LT, Ryman JC: MIRD: radionu-
clide data and decay schemes. New York, Society of Nuclear
Medicine, 1989.
 4
CHAPTER Single-Photon Emission
Computed Tomography
(SPECT) and Positron
Emission Tomography
(PET)

Radionuclide Tomography
SPECT RADIONUCLIDE TOMOGRAPHY
Instrumentation
Image Acquisition Conventional or planar radionuclide imaging suffers
Collimator Selection a major limitation in loss of object contrast as a result of
Orbit background radioactivity. In the conventional planar
Arc of Acquisition,Angular Sampling, and Matrix Size image, radioactivity underlying and overlying an object is
Imaging Time superimposed on radioactivity coming from the object.
Patient Factors The fundamental goal of tomographic imaging systems is
Image Reconstruction a more accurate portrayal of the distribution of radio-
Spatial Domain activity in the patient, with improved definition of image
Fourier Transformation and Frequency Domain detail. The Greek tomo means “to cut”; tomography may
Angular Projection (View) be thought of as a means of “cutting” the body into dis-
Projection Profile (Slice Profile) crete image planes. Tomographic techniques have been
Ray Sum developed for both single-photon and positron tomography.
Nyquist Frequency Rectilinear scanners with focused collimators repre-
Backprojection sent a crude type of tomography; the count rate sensitiv-
Filters ity is greatest in the collimator focal plane, and therefore
Reconstruction in the Frequency Domain more weight is given to radioactivity arising in that plane
Other Reconstruction Techniques than in planes superficial or deep to it. However, this is
Attenuation Correction not “true” tomography because the blurred out-of-plane
Image Reformatting:Transaxial, Sagittal, Coronal and Oblique Views activity contributes to the image.
Quality Assurance Restricted angle or longitudinal (frontal) tomography
PET shares the phenomenon of the rectilinear scanner:
Instrumentation inplane data are kept in focus, with blurring of out-of-
Gantry Size plane data. Restricted angle or longitudinal tomography
Detector Materials is analogous to conventional x-ray tomography, in which
Coincidence Detection the relative positions of the film and x-ray source remain
Spatial Resolution constant for the desired image plane but move relative to
Image Reconstruction each other in the overlying and underlying planes, blur-
Attenuation Correction and Quantitative Analysis ring the out-of-plane structures. A number of restricted
PET-CT angle systems were in vogue in the late 1970s and early
Comparison of PET and SPECT 1980s,including seven-pinhole collimator systems,pseudo-
SPECT Camera Imaging of 511 keV Positrons random coded aperture collimator systems, and various
(SPECT-PET) rotating slant-hole collimator systems.
Rotating gamma camera tomographic systems offer the
ability to perform true transaxial tomography. The most
important characteristic is that only data arising in the

52
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET)
image plane are used in the reconstruction or creation of
the tomographic image. Rotational single-photon emis-
sion computed tomography (SPECT) shares this feature
with x-ray computed tomography (CT) and positron emis-
sion tomography (PET). This is an important characteristic
because it offers a higher image contrast than with tomo-
graphic systems,which merely blur the out-of-plane data.
SPECT
With use of conventional radiopharmaceuticals, SPECT
allows true three-dimensional (3D) image acquisition and
display. Reconstruction of cross-sectional slices has tradi-
tionally used filtered backprojection, the same methodol-
ogy used for CT. However, newer systems offer iterative
approaches to image reconstruction.
Instrumentation
The most common approach to rotational SPECT is to
mount one or more gamma camera heads on a special
rotating gantry. Original systems used a single head, but
systems with two, three, and even four heads have been
developed. Today, two-headed systems are the most com-
mon commercially available SPECT systems. In particu-
lar, two-headed systems that allow flexibility in
orientation between the heads have become popular.
For body imaging, the heads are typically arrayed parallel
to each other; for cardiac applications, they are often
placed at right angles (Fig. 4-1)
Multiple heads are desirable because they allow more
data to be collected in a given period. Rotational SPECT
is “photon poor” compared with x-ray CT. Therefore, it is
desirable to collect as many counts as possible and com-
plete imaging within a reasonable time because of radio-
pharmaceutical pharmacokinetics and limits of the
patient’s ability to remain still. Thus,a study of Tl-201 dis-
tribution in the heart should be accomplished before sig-
nificant redistribution occurs.
Two camera heads
180 apart
(parallel)
Gantry
Figure 4-1 Two configurations for dual-headed SPECT systems.
53
In addition to the special gantry that permits camera
head rotation, modifications have been necessary for
rotational SPECT. Photomultiplier tube performance can
be affected by gravitational and magnetic fields. These
change depending on rotational angle, and subtle alter-
ations in photomultiplier tube energy response can
degrade images. Magnetic shielding of photomultiplier
tubes reduces this problem.
Rotational SPECT has highlighted the need to improve
every aspect of gamma camera system performance.
Flood field nonuniformities are translated as major arti-
facts in tomographic images because they distort the data
obtained from each view or projection. Desirable charac-
teristics for SPECT are an intrinsic spatial resolution (full
width at half maximum [FWHM]) of 3 mm, linearity dis-
tortion of 1 mm or less, uncorrected field uniformity
within 3%, and corrected field uniformity within 1%.
All contemporary rotational SPECT systems have online
uniformity and energy correction. Nonlinearities in photo-
multiplier tube energy response degrade both gamma cam-
era energy resolution and spatial resolution. Degraded
energy resolution is devastating because 35% or more of
recorded events can represent Compton-scattered pho-
tons. Poor energy resolution degrades the ability to reject
scattered photons on the basis of pulse height analysis. It
also degrades spatial resolution through decreased accu-
racy of determining x and y event localization coordinates.
Image Acquisition
Box 4-1 summarizes factors that must be considered in
performing rotational SPECT. In addition to standard
gamma camera quality control, confirmation is needed
that the axis of rotation corresponds to the center of the
matrix in the computer. Incorrect alignment results in
a blurring of the image or poorer resolution.
Collimator Selection
Collimator selection is generally limited to those sup-
plied by the system vendor. As discussed previously, for
Two camera
heads
90 apart
(perpendicular)
54 NUCLEAR MEDICINE: THE REQUISITES

Box 4-1 Image Acquisition Issues Camera head

for Single-Photon Emission
Computed Tomography

Center of rotation check

Collimator selection
Energy window selection
Orbit
Matrix size
Angular increment—number of views
180- vs. 360-degree rotation
Table
Time per view
Total examination time
Patient factors
Circular
orbit
a given septal thickness and hole diameter, collimators
with longer channels have higher resolution and lower
sensitivity. Even though SPECT is relatively photon poor, Camera head
collimator selection should favor higher resolution
whenever possible. This means selecting the high-resolu-
tion collimator over a high-sensitivity or general purpose
collimator for studies using Tc-99m. The multiheaded
systems permit the operator to trade the improved count
rate sensitivity for improved resolution by using ultra-
high-resolution collimators.
Special collimator options are available for imaging
the brain. Fan-beam and cone-beam collimators permit
Table
more of the camera crystal to be used for radiation
detection. The fan- and cone-beam collimators are simi- Elliptical
lar in geometry to converging collimators. They cause orbit
magnification of the object being imaged when it is Figure 4-2 Circular orbit (top) and elliptical orbit (bottom).
placed proximal to the focal point of the respective col-
limators. The resulting geometric distortion of the acquisition is typically a full 360 degrees. For studies
image data must be taken into account during image with Tc-99m-labeled agents it is often feasible to use a
reconstruction. high-resolution collimator and to acquire data in a 128 ×
128 matrix with an angular increment of 3 degrees or a
Orbit total of 120 angular projections. If these studies are per-
The orbit selected depends on the organ of interest and formed with a general purpose or lower resolution colli-
whether the system being used offers a noncircular orbit mator, a 64 × 64 matrix is typically selected with an
capability. The ideal orbit keeps the gamma head as close angular sampling increment of 6 degrees for a total of 60
to the organ of interest as possible because, for parallel- angular projections or 4 degrees for a total of 90 angular
hole collimators, the resolution is best at the face of the projections. These combinations of sampling increment,
collimator. Most imaging is still done using circular orbits matrix size, and collimator selection “balance” the resolu-
but contemporary systems permit the use of customized tion of the respective parameters. For imaging of tumors
noncircular orbits, which better approximate body and infections with gallium-67 (Ga-67) or indium-111 (In-
contours (Fig. 4-2). Orbit selection attempts to minimize 111) tracers, a 64 × 64 matrix is selected with a 6-degree
the distance between the camera head and the object angular sampling increment. (Note that in some SPECT
being imaged. systems, 64, 96, or 128 angular projections are used. For
simplicity, this discussion refers to 60, 90, and 120 projec-
Arc of Acquisition, Angular Sampling, tions only.)
and Matrix Size The merits of 180- vs. 360-degree rotation for cardiac
The choice of angular sampling interval and arc of acqui- studies have been debated in the literature. A minimum
sition depend on the clinical application and collimator arc of 180 degrees is necessary for true transaxial tomog-
selection. For body imaging applications, the arc of raphy. Proponents of the 180-degree approach argue that
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET)
Camera head
45 RAO
Table
135 LPO
Figure 4-3 The 180-degree arc frequently used for cardiac
imaging.
because the heart is close to the anterior chest wall, the
best data are available by imaging in a 180-degree arc typ-
ically spanning 135-degree left posterior oblique to
45-degree right anterior oblique (Fig. 4-3). The use of
180-degree arcs for cardiac SPECT is widely accepted in
clinical practice, particularly for cardiac imaging when
attenuation correction is not used.
Another question is whether to use continuous data
acquisition or “step-and-shoot” acquisition. Continuous
acquisition has the advantage of not wasting time during
movement of the camera head from one angular sampling
position to the next. However, the data are blurred by
the motion artifact of the moving camera head. The result-
ing tradeoff between sensitivity and resolution favor
step-and-shoot acquisition for most clinical applications.
Exceptions are applications with rapidly changing tracer
distribution and when determination of overall tracer
concentration is more important than spatial resolution.
Imaging Time
Most clinical protocols require a total imaging time of
20–40 minutes. Correspondingly, the time per projec-
tion is usually 15–30 seconds, but as much as 40–60 sec-
onds may be needed for relatively photon-poor studies
with Ga-67 and In-111.
Patient Factors
A major limitation in data acquisition time is the ability of
the patient to remain still throughout the imaging proce-
dure. Within accepted limits for dosimetry and radiation
exposure, a larger administered dosage allows for more
available counts. Although clinically accepted limits for
administered radioactivity should never be exceeded, the
radiation risk vs. benefit must take into account the likeli-
hood of obtaining a diagnostic quality image. The goal of
obtaining higher counting statistics is meaningless if the
patient moves,causing data between the different angular
sampling views to be misregistered. Currently available
55
Figure 4-4 SPECT artifact caused by injection site activity.
Degraded SPECT image of the liver and spleen caused by including
activity at the injection site in the imaging field of view. The
starburst artifact is due to backprojection of the hot spot activity
across the image. In this case, the degree of activity in the injection
site could not be accommodated in the reconstruction algorithm.
motion correction programs correct the data in one
dimension (vertical motion) but not three dimensions.
Patient compliance is improved by taking time during
setup to position the patient comfortably. For scans of
the head,the patient’s arms can be in a natural position at
the sides. For rotational SPECT studies of the heart, tho-
rax, abdomen, or pelvis, the arms are typically raised out
of the field of view so that they do not interfere with the
path of photons toward the detector. In all applications,
it is important to keep the injection site out of the field of
view to prevent artifacts resulting from residual or infil-
trated activity (Fig. 4-4).
Image Reconstruction
Each commercially available SPECT system takes a some-
what different and proprietary approach to the image
reconstruction process. Filtered backprojection has been
the standard method for SPECT reconstruction.
However, newer systems now offer iterative approaches
for reconstruction. Reconstruction is accomplished
either in the spatial domain or in the frequency domain
after Fourier transformation of the raw data. All
approaches to reconstruction use mathematical filters
that alter the raw data to facilitate tomographic image cre-
ation. Although reconstruction of SPECT images is highly
analytical, it is also an art. Different observers prefer differ-
56 NUCLEAR MEDICINE: THE REQUISITES
Figure 4-5 Frequency graph of image profile data. The graph corresponds to the cursor in the
image on the left after Fourier transformation. The frequencies are scaled as a fraction of Nyquist.
Their frequencies could also be scaled in terms of cycles per pixel or cycles per centimeter. Nyquist
1.00 corresponds to 0.5 cycle/pixel.
ent characteristics in the final images that are determined
by operator-adjustable parameters,including filtering.
Before a discussion of the image reconstruction
process, the following terms should be defined.
Spatial Domain
The spatial domain is the one in which we live. Its ter-
minology is that of counts per pixel, and measurements
of pixel size are in millimeters or centimeters.
Fourier Transformation and Frequency Domain
The French mathematician Fourier demonstrated that
any continuous function,such as projection profiles,in the
spatial domain can be approximated within an arbitrarily
determined value by the sum of a series of trigonometric
functions of varying frequencies and amplitudes. This
process is known as Fourier transformation. After
Fourier transformation, the data are said to reside in the
frequency domain, reflecting the periodicity of trigono-
metric functions. Figure 4-5 is a frequency graph of
image data after Fourier transformation. One cycle of
a periodic function is the interval from peak to peak. High-
frequency phenomena have short cycles and vary rapidly,
and low-frequency phenomena have longer cycles. The
distance between maximum and minimum values in a
periodic function is termed the amplitude (Fig.4-6).
One way to consider the Fourier transform is that it is
a plot of the amplitude as a function of frequency of the
trigonomic functions (sines and cosines). When added,
these functions yield the spatial domain representation
of interest. In the frequency domain, low frequencies
yield the overall shape of the object and high frequen-
cies yield the sharp corners and fine detail associated
with the object.
One cycle
Amplitude
Figure 4-6 A periodic function. One cycle is the distance from
peak to peak. Amplitude is the distance from peak to trough.
The advantage of working with image data in the fre-
quency domain is the relative simplicity of the mathemat-
ical manipulations once the data have been transformed.
Less computing power and computational time are
required than to perform reconstructions on the raw data
in the spatial domain. As computing power is becoming
less expensive, this relative advantage is fading.
Angular Projection (View)
The term angular projection (or view) refers to the stan-
dard planar images obtained at each angle of SPECT
acquisition. The SPECT raw data set typically has 60–120
angular projections, corresponding to angular increments
between 6 degrees and 3 degrees, respectively. Figure 4-7
illustrates the detector in two sampling positions.
Projection Profile (Slice Profile)
The angular projections exist in the computer as either
64 × 64 or 128 × 128 matrices. A projection profile, also
referred to as a slice profile, represents the data in one
row of the matrix. The raw data for a given tomographic
slice come from all the projection profiles corresponding
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) 57

cycle/pixel). The Nyquist frequency can also be

expressed in cycles per centimeter. Thus, for an acquisi-
tion matrix with 6 mm pixels, the Nyquist frequency
would equal approximately 0.8 cycle/cm (0.5 cycle/pixel
= 0.5 cycle/0.6 cm = 0.8 cycle/cm). The Nyquist fre-
quency is an important consideration in the design and
Detector in selection of filters used in the tomographic reconstruc-
two sampling tion process.
Point source positions
of radioactivity
Backprojection
The concept of backprojection is fundamental to the
reconstruction of tomographic images from the raw
data. Backprojection takes the line data from the pro-
jection profiles and projects it back into a two-dimen-
sional (tomographic) image. In simple backprojection in
the spatial domain, the count values or ray sums in each
pixel of the projection profiles corresponding to
Figure 4-7 Acquisition positions. Point source of radioactivity a given tomographic slice are first redistributed equally
with the detector illustrated in two sampling positions. Typically
along the corresponding rays (see Fig. 4-8). The distribu-
60–120 sampling positions are used for SPECT.
tion is equalized along the ray because there is no way
of telling from what depth the counts originated.
These recorded values for each ray from all sampling
angles are added together at their intersections in
Backprojections
the tomographic image plane. That is,at each pixel in the
tomographic image plane, rays from all of the angular
Image profiles projections intersect, and the count value given to the
Intersection of of point source pixel is the sum of the values assigned to all the rays
backprojected
rays at location intersecting at that point (see Fig. 4-8). Hot spots are
of point source associated with high count values in the backprojected
rays intersecting at their corresponding location. Cold
spots do not contribute to counts in the individual ray
projections and the cumulative value of the corre-
sponding summation is less.
Figure 4-8 Backprojection for two rays obtained at different If one performs only simple backprojection, satisfac-
sampling angles.The respective counts for the rays are projected tory tomographic images are not obtained. Reconstructing
for each pixel along their paths. Note the summation at the point a point source results in a “star” artifact with exaggerated
of intersection. borders of the point source itself and starburst ray artifacts
emanating from it (Fig. 4-9). In the frequency domain, in
to that slice in the angular projection views. Thus, a study which reconstruction is performed, a mathematical
with 60 angular projections yields 60 projection profiles (ramp) function is applied to the projection data that
as the input data for reconstruction of each tomographic allows for tomographic reconstruction.
image (Fig. 4-8).
Filters
Ray Sum To solve the problem of the star artifact that arises from
The value of each pixel in a projection profile is called simple backprojection and to address issues of back-
the ray sum. It is equal to the total activity recorded ground and noise, image data are “filtered.” The filters are
from the corresponding ray perpendicular to the camera mathematical functions designed for enhancement of
face in the plane of interest. desired characteristics in the image. These include elimi-
nation of the star artifact, background subtraction, edge
Nyquist Frequency enhancement, and suppression of statistical noise by
The Nyquist frequency is the highest frequency that can selectively emphasizing certain frequency components of
be resolved in the image, based on the resolution charac- the image. A good analogy is the audio graphic equalizer
teristics of the imaging system and the parameters that allows one to selectively enhance or minimize cer-
selected for data acquisition. For SPECT, the Nyquist fre- tain frequency components in order to improve the qual-
quency is equal to 0.5 cycle/pixel (1 Nyquist = 0.5 ity of the audio signal. Spatial filters can work in much
58 NUCLEAR MEDICINE: THE REQUISITES

1.0

Ramp

Amplitude
0.5

Cut-off
frequency

0.25 0.5
Frequency
(cycles/pixel)

1.0

Hamming

Amplitude
0.5

Figure 4-9 Star artifact resulting from simple backprojection.

The star results from the multiple summations in the areas of
intersection of the backprojected rays.
0.25 0.5
the same way, allowing the user to improve the image Frequency
(cycles/pixel)
quality by enhancing and minimizing certain frequencies.
1.0
Low-pass filters selectively let through low frequen-
cies and filter out high frequencies in the data; the oppo-
site applies for high-pass filters. Background activity,
Butterworth
Amplitude

including the star artifact, resides in the low-frequency

portion of the spectrum. Statistical noise exists at all fre-
quencies but becomes dominant at higher frequencies.
Thus, using a high-pass filter will improve the fine detail
of the image but may also lead to more image noise.
In diagrams of filter functions in the frequency
domain,the amplitude is plotted on the y-axis and the fre-
quency is plotted on the x-axis (Fig. 4-10). The frequen-
cies and amplitudes under the filter function are “passed”
by the filter. Since Fourier series are by definition infi-
nite, a cutoff frequency is also defined for practical pur-
poses and is typically equal to the Nyquist frequency.
This makes sense because frequencies higher than the
Nyquist frequency cannot contribute additional informa-
tion to the image. Restricting the filter function to a cut-
off frequency simplifies the calculations.
As the name implies, the ramp filter has the shape of
a straight line extending up from the origin when
graphed in frequency space (see Fig.4-10). By inspecting
the area under the curve,one can see that ramp filters are
high-pass filters. Ramp filters are applied in filtered
backprojection reconstruction algorithms to suppress
the star artifact and low-frequency noise. They also elimi-
nate low frequencies from the signal.
The ramp filter takes care of the star artifact and low-
frequency background, but other filters are used to sup-
0.5
Cut-off frequency
0.25 0.5
Frequency
(cycles/pixel)
Figure 4-10 Ramp, Hamming, and Butterworth filters. The
ramp filter is a “high-pass”filter designed to reduce background
activity and the star artifact. Hamming and Butterworth filters are
“low-pass”filters designed to reduce high-frequency noise.
press high-frequency noise. Many of these filters have
been named after their inventors, and such names as
Butterworth, Hamming, Hanning, and Hann are frequently
seen in the literature (see Fig. 4-10). These low-pass filters
eliminate higher frequency noise components. Too little
filtering of high-frequency noise results in images with
excessively grainy texture. Too much filtration of high-
frequency data results in oversmoothing of images with
loss of edge definition (Fig. 4-11). Areas where radioactiv-
ity concentrations change rapidly, such as the borders of
organs, are represented in the high-frequency data, and
oversmoothing blurs borders.
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) 59

space. A ramp filter is applied to the transformed pro-

files. Then smoothing and edge enhancement/preserva-
tion filters are applied. The filtered profiles are then
summed from all projection angles akin to backprojec-
tion. Finally, an inverse Fourier transform is applied to the
data to create the reconstructed image in the spatial
domain. Alternatively, the inverse Fourier transforma-
tion can be performed after filtration and the back-projec-
tion accomplished in the spatial domain.

Other Reconstruction Techniques

With the availability of increasing computer power,
Figure 4-11 Effects of different filters on the appearance of
SPECT liver and spleen images. A, The filter has resulted in iterative methods are increasingly used for the recon-
excessive noise texture in the image. B, The filter has struction of SPECT and other tomographic images. In
oversmoothed the image, with loss of detail. the iterative approach, an initial set of tomograms is
reconstructed. This 3D data set is then used to create
The Butterworth filter is commonly used because it a new set of reprojection images. If the reconstruction
both smoothes noise and preserves edges. It is particu- process were perfect, the reprojection images would
larly flexible because it allows the operator to select two be identical to the initial projection images. Because
defining parameters, the cut-off frequency and the order. this is not the case, the difference between the original
The cutoff frequency is sometimes called the power of projection and the reprojection images based on
the filter and, as described previously, is the maximum tomographic reconstruction is used as input for
frequency that a filter will pass. For the Butterworth fil- another iteration of reconstruction. This process can
ter, the order is a parameter that controls the shape or go on as long as is practical or until there is no further
slope of the filter. convergence between the reprojection views based
on the tomographic data and the initial projection
Reconstruction in the Frequency Domain images.
With the foregoing concepts in hand, it is possible to Each iteration takes about as long as one implementa-
describe the entire reconstruction process as it applies tion of filtered backprojection. Because early algorithms
to filtered backprojection in frequency space or the fre- required tens if not hundreds of iterations to achieve an
quency domain (Fig.4-12). First,the individual projection acceptable image, these were considered too slow for
profiles undergo Fourier transformation into frequency routine clinical use. However, with newer, more efficient

1.0 1.0
Amplitude

Amplitude

Fourier 0.5
Ramp 0.5

Projection profiles transformation filter

from projection images:
spatial domain
0.25 0.5 0.25 0.5
Frequency (cycles/pixel) Frequency (cycles/pixel)

Image data in
frequency space

1.0

Inverse Fourier
Amplitude

0.5 transform and

Butterworth backprojection
filter of filtered
projection
0.25 0.5
profiles Tomographic image in
Frequency (cycles/pixel) spatial domain
Figure 4-12 Steps in filtered backprojection reconstruction for SPECT.
60 NUCLEAR MEDICINE: THE REQUISITES
algorithms and more powerful, faster computers, itera-
tive reconstruction can now be performed in a reason-
able amount of time for routine clinical use.
Ordered subset expectation maximization (OSEM) is
a commonly used form of iterative reconstruction that
speeds the process using limited subsets of data. Iterative
reconstruction techniques offer the flexibility to include
corrections for system performance (e.g., scatter correc-
tion and resolution degradations) and are finding use in
various approaches to attenuation correction.
Attenuation Correction
A special problem of SPECT imaging is the attenuation of
radioactivity in tissue. Photons emitted from deeper
within the subject are more likely to be absorbed in the tis-
sue than those emitted from the periphery. Therefore, the
signals from these tissues are “attenuated.” To obtain an
image where the signal is not depth dependent, one must
therefore perform an attenuation correction. Evidence is
increasing that certain studies, such as myocardial perfu-
sion imaging, benefit from attenuation correction. There
are two fundamentally different approaches to the prob-
lem. Both are designed to create an image attenuation cor-
rection matrix, where the value of each pixel represents
the correction factor that should be applied to the corre-
sponding data in the reconstructed image.
For solid organs such as the liver, in which an assump-
tion of near uniform attenuation can be made, an analyti-
cal or mathematical approach such as the Chang
algorithm can be used. After the object is initially recon-
structed, an outline of the body part is made on the com-
puter for each tomographic slice. From this outline, the
depth and therefore the appropriate correction factor for
each pixel location can be computed.
The theoretical attenuation coefficient for Tc-99m in
soft tissue is 0.15 per centimeter. (This applies only to
“good” geometry, that is, a point source with no scatter
into the ray. The observed value for Tc-99m in the
abdomen is 0.12 per centimeter and in the brain is 0.13
per centimeter.) Thus, at a depth of 7 cm in a liver SPECT
study, almost 60% of the corresponding activity is attenu-
ated. The observed count value would have to be multi-
plied by a factor of 2.5 (0.4 × 2.5 = 1) to correct for
attenuation.
The major limitation of the analytical approach occurs
when multiple types of tissue,each with a different atten-
uation coefficient, are in the field of view. Cardiac imag-
ing is the most important example. The soft tissues of the
heart and thorax are surrounded by the air-containing
lungs and the bony structures of the thorax. To correct
for nonuniform attenuation, a transmission scanning
approach is used for attenuation correction. In essence,
a CT scan of the thorax is obtained using a radionuclide
source rather than an x-ray tube. Innumerable specific
techniques have been described that use sheet sources
of radioactivity, moving line sources, and arrays of line
sources.
The transmission SPECT scan can be obtained either
separately or simultaneously with the diagnostic SPECT
scan. In the simultaneous approach, a radionuclide such
as gadolinium-153 (Gd-153) or cobalt-57 (Co-57) is used
with a separate energy window set for the appropriate
photopeak. The high-energy photopeak of Gd-153 is
roughly 100 keV and the energy of Co-57 is 122 keV. For
studies using Tc-99m, correction for crosstalk caused by
downscatter from the 140-keV photons is done first, and
then the radionuclide transmission CT image is recon-
structed using the kinds of SPECT reconstruction tech-
niques described previously. This image is then
normalized and scaled for the difference between the
energy of the transmission source and the 140-keV photon
energy of Tc-99m. The resulting image is an attenuation
map of the thorax that can be applied pixel by pixel to
correct for the effects of attenuation.
It is hoped that this approach will address two linger-
ing problems with cardiac SPECT. For men, scans often
show decreased activity in the inferior wall, possibly
resulting from attenuation by overlapping organs
beneath the diaphragm. For women, overlying breast tis-
sue can significantly distort SPECT data by differential
attenuation. Both of these can lead to loss of signal from
certain portions of the myocardium. Appropriately
applied attenuation correction can correct for this and lead
to a more accurate diagnosis.
Another approach to nonuniform attenuation correc-
tion is the use of hybrid SPECT-CT imaging systems. In
these systems, the SPECT camera is interfaced with a CT
scanner. The image from the CT scans can thereby be
used as the transmission image used for attenuation cor-
rection. The CT scan can also be used for anatomical
correlation of the functional SPECT data. Several of the
SPECT camera manufacturers have recently introduced
such hybrid devices.
Image Reformatting: Transaxial, Sagittal,
Coronal, and Oblique Views
A particular advantage of gamma camera rotational
SPECT is that a volume of image data is collected at one
time. This permits the acquisition of multiple tomo-
graphic slices simultaneously and the registration of the
data between planes. Interslice filtering is also used to
reduce artifacts in reformatted data. In addition to the
standard transaxial images, other image planes that have
special relevance to the organ of interest can be recon-
structed. For example, sagittal and coronal images can be
directly generated from the transaxial images.
The resorting or reformatting approach is particularly
valuable in cardiac imaging (Fig. 4-13). The orientation
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) 61

Figure 4-13 Cardiac SPECT images reformat data into multiple planes. The top two rows are
short-axis views obtained perpendicular to the long axis of the left ventricle. The middle two rows
are horizontal long-axis images, and the bottom two rows are vertical long-axis images. The patient
has a large fixed perfusion defect involving the inferior wall of the left ventricle. The ability to
reformat the data allows more precise and accurate localization of abnormalities.

of the heart varies among patients. The heart usually has
a horizontal orientation in shorter subjects and a more
vertical orientation in taller ones. Image planes both per-
pendicular and parallel to the long axis of the heart
would be more useful. This is readily accomplished with
a volume data set. The computer operator defines the
geometry of the long axis of the heart. The computer is
programmed to resort the data to create cardiac long-axis
and short-axis planes oblique to the transaxial slices. The
optimum angulation is highly variable among patients,
reflecting the differing orientation of the heart.
A useful strategy is to reproject the tomographic data as
a sequence of planar images having the same fields of view
as the original angled sampling images. Viewing the recon-
structed projection images in cinematic mode gives an
excellent 3D display of the data. An additional advantage of
using the reconstructed data is that overlying structures
can be removed before the data are reprojected and
selected features in the data can be emphasized. For exam-
ple, in Tc-99m pyrophosphate imaging of the heart, the
ribs can be subtracted from the 3D data set before the data
are reprojected. The ribs no longer obscure the cardiac
activity. Another advantage of using reprojection rather
than tomographic images is that this technique provides a
better overall orientation of the heart in the chest.
A maximum intensity projection scan (MIPS) can be
created by reprojecting the hottest point along each par-
ticular ray for any given projection. These MIPS images
emphasize areas of abnormally increased accumulation
while providing a better overall orientation of the abnor-
mality to the skeleton than individual tomographic slices.
In some cases, the MIPS images are distance weighted to
make activity that is further from the viewer appear less
intense, thereby enhancing the 3D effect. Looking at
individual transaxial tomograms can be confusing with-
out knowing a lesion’s location relative to surrounding
structures.
Quality Assurance
The projection data from all SPECT scans should be
inspected before image reconstruction. Excessive
patient motion degrades the quality of SPECT scans
because of misregistration of data in the different angular
projections. Patient motion can be assessed in a number
of ways. When the angular unprocessed projections are
viewed in a cinematic closed loop display, excessive
patient motion is readily detected as a flicker or disconti-
nuity in the display. Some laboratories use radioactive
marker sources placed on the patient to assess motion.
62 NUCLEAR MEDICINE: THE REQUISITES

Another approach is to view a sinogram of a slice.

Sinograms are constructed by placing the projection pro-
files for a given tomographic slice in a stack. The borders
of the sinogram should be smooth and interslice changes
in intensity should be small; any discontinuity indicates
motion of the patient (Figs. 4-14 and 4-15). Only up-and-
down motion can be corrected.
Rotational SPECT requires maximum performance
from the gamma camera. All standard quality control pro-
cedures are observed, as well as several additional points
(Box 4-2). The alignment of the detector, gantry, and
imaging table is critical. In transaxial rotation, the basic
assumption is that the face of the collimator is truly paral-
lel to the axis of rotation. If it is off-axis, the tilted field of
view of the collimator will result in misregistered data.
Similarly for multihead cameras, the detector heads must
be aligned with each other for correct registration of
data. Another fundamental assumption is that the center
of rotation corresponds to the center of the image matrix
in the computer. If the center of rotation is offset, it man-
ifests as degradation in resolution.
The gamma camera–computer interface is particu-
larly important in rotational SPECT. The pixel size must
be carefully calibrated. Attenuation correction depends
on depth estimates, and a change in pixel size will Figure 4-15 Sinogram illustrating multiple gaps in the
sequential profile data. Compare these discontinuities with the
change distance and therefore attenuation correction regular progression of data in Figure 4 -14. The discontinuities
factors. Pixel size is adjusted by the setting of the ana- indicate unwanted motion of the object from one sampling
log-to-digital converters and should be checked in both position to the next.

Figure 4-14 Sinogram from a myocardial perfusion study. The sinogram corresponds to the level
of the cursor in the image on the left. Note the regular progression in the data across the projection
profiles, indicating stability and lack of unwanted movement of the heart from one projection view
to the next.
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) 63

the x and y dimensions. Pixel width should be identical

in the x and y directions. The y-axis determines slice
thickness, and a difference in x and y pixel dimensions
will create problems in reformatting image data into
oblique planes. A shift in the performance of the ana-
log-to-digital converters can result in movement of the
center of rotation.
Field uniformity corrections are critical in SPECT
imaging. Detector nonuniformity results in bull’s-eye or
ring artifacts. The usual count flood image of 1 million
to 5 million obtained for planar imaging is inadequate for
uniformity correction in SPECT imaging. For large field-
of-view cameras and a 64 × 64 matrix, 30 million counts
are acquired (roughly 10,000 counts per pixel) in the
image to achieve the desired relative standard deviation
of 1%. For a 128 × 128 matrix, four times this amount or
120 million counts must be acquired to achieve the same
level of statistical precision.
Acquiring this number of counts requires a significant
amount of time. The temptation to use very large
amounts of radioactivity should be avoided because high
count rates can also result in degraded performance of
gamma camera electronics and in the recording of spuri-
ous coincidence events. Conservatively, the correction
floods should be obtained at 20,000 counts per second,
or less. The radioactivity in the flood itself must have
a uniformity within 1%. Water-filled sources are subject
to problems of incomplete mixing and introduction of
air bubbles, as well as bulging of the container. For these
reasons, Co-57 sheet sources are more convenient and
more reliable than water-filled sources.
PET
PET is a form of tomography made possible by the
unique fate of positrons. When positrons undergo anni-
hilation by combining with negatively charged electrons,
two 511-keV gamma rays are given off in opposite direc-
tions 180 degrees apart. In contrast to SPECT imaging,
which detects single events, in PET imaging two detector
elements on opposite sides of the subject are used to
detect paired annihilation photons. If the photons are
detected at the same time (or “in coincidence”), the
event is assumed to have occurred along the line con-
necting the two detectors involved. Thus, the direction
of the photons can be determined without the use of
absorptive collimation.
Coincidence detection leads to at least a 100-fold
increase in the sensitivity of PET relative to conventional
nuclear medicine imaging and explains the higher qual-
ity of images as compared to SPECT. The counts occur-
ring between a single pair of detectors can be
considered a ray sum just as in SPECT, and thereby pro-
jections can be generated that can be reconstructed just
as in SPECT. Both filtered backprojection and iterative
approaches such as OSEM can be used to reconstruct
the data (Fig. 4-16).
Instrumentation
Instrumentation for PET has undergone several genera-
tions of development. Early systems had a single ring

BOX 4-2 Quality Assurance in Single- Paired

Photon Emission Computed detector
Tomography elements

Parameter Comment

Center of rotation Should match center of image

matrix in the computer; look Positron
for horizontal shift on x-axis emitted
Pixel size X and y dimensions should be Annihilation
equal; any change in pixel size Travel
requires recalibration of in tissue
attenuation correction factors
Uniformity Counts of 3 million for routine 511-keV
intrinsic and extrinsic photons
uniformity checks; 30 million 180ⴗ apart
for input for uniformity
correction
Spatial resolution Weekly per usual gamma camera
and linearity quality control
Detector head Camera face parallel to axis of Ring
alignment rotation detector
Head matching Alignment of multiple heads for Figure 4-16 PET ring detector. After emission, positrons travel
correct event localization a short distance in tissue before the annihilation event. The 511-keV
protons are given off 180 degrees apart.
64 NUCLEAR MEDICINE: THE REQUISITES
with multiple detectors and generated a single tomo-
graphic section at a time. Now, PET typically consists of
many rings of multiple detectors. Each detector in the
ring is typically paired with multiple other detectors
on the opposite side of the detector ring. These detec-
tors, or the arc of detectors, are selected to encompass
the field of view of the object or organ being imaged
(Fig. 4-17).
Multiple-ring systems were rapidly developed, allow-
ing a volume to be imaged simultaneously. Early systems
with three to eight rings of detectors typically had septa
inserted between the tomographic planes to shield the
detectors from crosstalk from activity arising outside of
the plane of interest. These multiple-ring systems with
septa inserted between the tomographic planes are often
referred to as two-dimensional systems.
The technical development of PET instrumentation
has now reached the point where systems have as many
as 32 rings of detectors with the capability of creating
a simultaneous tomographic section for each ring and an
additional section between each pair of rings, for a total
of 63 simultaneously acquired tomographic images.
Contemporary systems have retractable septa between
the planes and are referred to as 3D systems. This design
greatly increases system sensitivity. However, with larger
patients, scatter and the presence of random coinci-
dences can limit the quality of the 3D PET images. The
concept of pairing each detector with multiple detectors
on the opposite side of the ring has been retained with
these systems and extended to pairings between differ-
ent rings.
Object
Figure 4-17 Pairing of detectors. In the PET tomograph, each
detector is paired with multiple detectors on the opposite side of
the ring to create an arc encompassing the object. This multiple-
pairing strategy increases the sensitivity of the device.
Gantry Size
Similar to the history of x-ray, CT, and magnetic reso-
nance imaging, the first PET scanners were designed
for head imaging. The early PET systems had a typi-
cal diameter of 60 cm. Current systems are suitable
for head and body imaging, with a typical diameter of
100 cm.
Detector Materials
The density and effective atomic number (Z) for NaI(Tl)
crystals are not ideal for “stopping” or detecting the
511-keV gamma rays used in PET imaging. Bismuth ger-
minate oxide (BGO) is approximately twice as dense with
an effective Z of 74, compared with an effective Z of 50
for NaI(Tl). BGO detectors have been used extensively in
PET imaging applications for this reason. Other detector
materials that have found application include cesium
fluoride and barium fluoride. These have much faster
resolution than BGO but are not as dense (Table 4-1).
New detector materials, such as lutetium oxyorthosili-
cate (LSO) and gadolinium oxyorthosilicate (GSO),are cur-
rently in use in several state-of-the-art PET scanners. They
combine the high density of BGO with far better time res-
olution and superior light yield. This material shows
promise as the detector material of choice for the future.
Coincidence Detection
Special circuitry in the PET tomograph allows detection of
two gamma ray photons given off by a single positron
annihilation event. The two events are considered to be
from the same event if they are counted within the coinci-
dence timing window. The coincidence window is on the
order of 10 nsec. Thus, when events are registered in
paired detectors within 10 nsec of each other, they are
accepted as true coincidence events. If a recorded event is
not matched by a paired event within the coincidence
time window, the data are discarded. This approach
effectively provides “electronic collimation” to define the
Table 4-1 Characteristics of Positron Emission
Tomography Detector Materials
Delay
Material Density Effective Z time (nsec)
Bismuth germinate 7.13 74 300
oxide (BGO)
Gadolinium 6.71 59 60
oxyorthosilicate
(GSO)
Lutetium 7.40 66 40
oxyorthosilicate
(LSO)
Sodium iodide (NaI[T1]) 3.67 50 230
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) 65

tomographic image planes. By not having to physically
collimate the detector elements, PET tomographs offer
much higher sensitivity than would otherwise be the case.
One of the problems in the coincidence approach is
the presence of paired random events that appear to the
detection circuitry as paired annihilation photons (Fig.
4-18). Paired random events are two photons arising
from two different positron annihilation events and are
therefore not useful in reconstructing the true location
of tracer distribution. As the amount of radioactivity in

True No events:
coincidence both photons
event absorbed or
scattered out
of ring

Single event Misregistered

event

Absorption Compton
or scattering
scattering

False event:
coincident
random
Misregistered single
event: events
noncolinearity

Figure 4-18 Different possibilities in positron decay and event detection with PET. The wanted
event is a true coincidence event. Single events are easily rejected but contribute to processor dead
time. Misregistered events caused by noncolinearity are difficult to discriminate.“False events”may
be incorrectly accepted if the two photons are intercepted in paired detectors.
66 NUCLEAR MEDICINE: THE REQUISITES
the field of view increases and the count rate increases,
the number of falsely recorded paired random events
also increases.
Spatial Resolution
The spatial resolution of modern PET tomographs is
excellent. Specialized experimental devices approach
1.5-mm resolution ( FWHM ) as measured by a line source
in air. Resolution under clinical scanning conditions is
superior in PET compared with SPECT. Resolution for
clinical studies is in the 6- to 8-mm FWHM range with
high-end contemporary PET scanners.
The ultimate spatial resolution of PET is limited by
two physical phenomena related to positrons and their
annihilation. First, positrons are given off at different
kinetic energies. Energetic positrons such as those given
off in the decay of oxygen-15 (O-15), Ga-68, and rubid-
ium-82 (Rb-82) may travel several millimeters in tissue
before undergoing annihilation (Fig. 4-16). Thus, the
detected location of the annihilation event is some dis-
tance from the actual location of the radionuclide. This
travel in tissue degrades the ability to truly localize the
biodistribution of the radioactive agent in the patient.
The second phenomenon limiting resolution is the
noncolinearity of the annihilation photons. In addition
to the energy equivalent of the rest mass of two elec-
trons,the annihilation event incorporates residual kinetic
energies of the positron and the negative electron with
which it combines. This results in a small deviation from
true colinearity along a single ray (see Fig. 4-18). The
angle by which the gamma rays depart from the theoreti-
cal 180-degree colinearity results in a 1- to 2-mm spatial
uncertainty in event localization for clinical whole body
PET scanners.
Image Reconstruction
Image reconstruction in PET uses many of the same prin-
ciples as SPECT. Filtered backprojection and iterative
reconstruction algorithms have both found application.
In three-dimensional systems, crossplane information is
incorporated into the “in-plane”or “direct plane”data.
Attenuation Correction and Quantitative Analysis
A unique and important characteristic of PET is the abil-
ity to correct for attenuation of the 511-keV gamma rays
in tissue. The basis of this ability is the fact that attenua-
tion and therefore coincidence detection of positron
annihilation are independent of the location along
a given ray between opposite detectors. Because the
total amount of tissue traversed by the two photons is
a constant for each ray, the correction factor for each
coincidence line can be determined empirically by per-
forming a transmission scan. The observed count rate
obtained along each ray during the actual scan is cor-
BOX 4-3 Quantitative Measurements
by Positron Emission Tomography
Regional (absolute) radionuclide localization
pH
Blood flow
Blood volume
Oxygen extraction fraction
Oxygen metabolism
Glucose metabolism
Receptor binding and occupancy
rected by dividing it by the attenuation factor. This
approach to attenuation correction assumes that the
patient does not move between the transmission scan
and the emission scan. The transmission scan must have
sufficient counting statistics to avoid introducing statisti-
cal error into the data.
The ability to correct for attenuation improves the
quality of PET images and permits absolute quantifica-
tion of radioactivity in the body. Quantitative analysis is
the basis for numerous metabolic, perfusion, and biodis-
tribution measurements. For example,a therapeutic drug
can be radiolabeled with a positron-emitting radionu-
clide. With knowledge of the specific activity of the radi-
olabeled drug and the ability to correct for attenuation,
the absolute uptake and distribution of the drug can be
quantitatively measured. Box 4-3 summarizes several of
the important quantitative measurements used in appli-
cations of PET imaging.
PET-CT
The recent introduction of hybrid PET-CT scanners has
allowed for the direct correlation of the functional infor-
mation available from PET and the anatomical informa-
tion from CT. These devices place the CT scan directly in
front of the PET scanner. The helical CT scan is acquired
first, followed by the PET scan. The CT scan can then
provide both a transmission scan for attenuation correc-
tion as well as anatomical correlation. Because the CT
scan can be acquired much faster than a traditional PET
transmission scan, the use of PET-CT can substantially
increase patient throughput.
Artifacts can be introduced by the CT-based attenua-
tion correction caused by misregistration between the
two image sets. For example, the difference in breathing
patterns between the two scans can make it difficult to
register the two studies in the area of the diaphragm.
However, PET-CT has been extremely useful in anatomi-
cally defining both pathology and normal anatomy on
 Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET)
the PET scan. Areas of increased FDG uptake can be
more easily correlated with a metastatic lymph node or
shown to be associated with a region of brown fat or
intestine. For these reasons, the use of PET-CT has
increased dramatically in the few years and will most
likely continue to grow into the future.
COMPARISON OF PET AND SPECT
The advantages of PET are superior sensitivity and resolu-
tion and a far greater flexibility of incorporating positron
labels into biomolecules. PET scanners are considerably
more expensive than SPECT systems and also require the
presence of an onsite cyclotron for a full range of appli-
cations.
SPECT has significant cost advantages. SPECT systems
are smaller and easier to place within hospitals. SPECT
has a singular advantage in being applicable to the most
commonly performed procedures in nuclear medicine,
including myocardial perfusion imaging with either thal-
lium-201 (Tl-201) or Tc-99m and oncological imaging
with Ga-67 citrate.
SPECT CAMERA IMAGING OF 511 keV
POSITRONS (SPECT-PET)
The feasibility of imaging positron-emitting radionuclides
with SPECT systems has been widely explored. The use
of this approach has been established with greatest appli-
cability to studies of the heart and certain tumors using
fluorine-18 fluorodeoxyglucose (F-18 FDG).
Although initially imaged with very high-energy colli-
mators, noncollimator systems incorporating coinci-
dence detection circuitry have been developed for and
used with dual-headed SPECT devices. Spatial resolution
is considerably better when coincidence detection is
used. These systems became commercially available.
SPECT-PET was initially conceived as a way to bring
PET to those who did not have a dedicated PET system.
Gamma cameras were available in all nuclear medicine
clinics and this minimized the high initial cost of PET sys-
tems. However, with the rapid growth and acceptance of
67
dedicated PET systems, current SPECT-PET systems will
probably be replaced by dedicated PET systems and it
likely will be looked back upon as a temporary historical
bridge between SPECT and PET. On the other hand, it is
conceivable that systems may be developed in the future
that can perform both functions optimally.
SUGGESTED READING
Celler A, Sitek A, Stoub E, et al: Multiple line source array for
SPECT transmission scans: simulation, phantom and patient
studies, J Nucl Med 39: 2183–2189, 1998.
Chandra R: Nuclear medicine physics: the basics, 6th ed.
Baltimore,Williams & Wilkins, 2004.
Freeman LM, Blauflox MD: The coming age of PET (part 1).
Semin Nucl Med 28, 1998.
Freeman LM, Blaufox MD: The coming age of PET (part 2).
Semin Nucl Med 28, 1998.
Hichwa RD: Production of PET radioisotopes and principles of
PET imaging. In Henkin RE, editor: Nuclear medicine, St Louis,
1996, Mosby, pp 279-291.
Patton JA, Rollo FD: Basic physics of radionuclide imaging. In
Freeman and Johnson’s clinical radionuclide imaging,
3rd ed. Freeman LM, Ed. New York, Grune & Stratton, 1984.
Patton JA,Turkington TG: Coincidence imaging with a dual-head
scintillation camera, J Nucl Med 40: 4432-4441, 1999.
Phelps ME, Mazziotta JC, Schelbert HR: Positron emission
tomography and autoradiography: principles and applica-
tion for the brain and heart. New York, Raven Press, 1986.
Powsner RA, Powsner ER: Essentials of nuclear medicine
physics. Malden, MA, Blackwell Science, 1998.
Reivich M,Alavi A: Positron emission tomography. New York,
Alan R Liss, 1985.
Rollo FD: Nuclear medicine physics, instrumentation and
agents. St. Louis, Mosby, 1977.
Simmons GH: The scintillation camera. New York, Society of
Nuclear Medicine, 1988.
Cherry SR, Sorenson JA, Phelps ME: Physics in nuclear medi-
cine, 3rd ed. Philadelphia,WB Saunders, 2003.
Votaw JR: The AAPM/RSNA physics tutorial for residents:
physics of PET, Radiographics 15: 1179-1190, 1995.
Yester MV:Theory of tomographic reconstruction. In Nuclear
medicine. Henkin RE, Ed. St Louis, Mosby, 1996, pp 222-231.
 5
CHAPTER Endocrine System

Thyroid Scintigraphy and Function Studies Parathyroid Scintigraphy

Thyroid Anatomy and Physiology Anatomy and Embryology
Anatomy Pathophysiology
Physiology Hyperparathyroidism
Radiopharmaceuticals Radiopharmaceuticals
Radioiodine Tc-99m Sestamibi and Tc-99m Tetrofosmin
Tc-99m Pertechnetate Adrenal Scintigraphy
Which Radiopharmaceutical to Use Adrenocortical Scintigraphy
Special Considerations and Precautions Radiopharmaceuticals
Methodology for Thyroid Uptake Studies and Thyroid Scans Suppression Studies
Thyroid Uptake Normal Adrenocortical Scintigram
Thyroid Scintigraphy Cushing’s Syndrome
Clinical Indications for Thyroid Uptake Studies Hyperaldosteronism
Thyrotoxicosis Androgen Excess
Clinical Indications for Thyroid Scintigraphy Incidentalomas
Normal Thyroid Scintigraphy Adrenomedullary Scintigraphy
Abnormal Thyroid Scintigraphy Radiopharmaceuticals
Thyroid Nodule Technique
Goiter Precautions
Ectopic Thyroid Tissue Normal MIBG Scintigraphy
Subacute Thyroiditis Clinical Applications
Chronic Thyroiditis (Hashimoto’s Thyroiditis)
Acute Thyroiditis
Thyroid Cancer
Other Thyroid Imaging Radiopharmaceuticals THYROID SCINTIGRAPHY AND
Tl-201 and Tc-99m Sestamibi FUNCTION STUDIES
F-18 Fluorodeoxyglucose (FDG)
Iodine-131 Metaiodobenzylguanidine (MIBG) Thyroid studies were among the first nuclear medicine
Indium-111 Somatostatin Receptor Scintigraphy procedures. When iodine-131 became available to the
Other Thyroid Function Studies medical community in the United States after World War
T3 Suppression Test II, thyroidologists quickly recognized that the percentage
TSH Stimulation Test uptake of radioiodine at a fixed point in time after admin-
Perchlorate Discharge Test istration was a measure of thyroid function. This mea-
Radioiodine Therapy surement was further enhanced by suppression and
Graves’ Disease stimulation interventions aimed at determining thyroid
Toxic Nodular Goiter autonomy and thyroid functional reserve. By the early
Thyroid Cancer 1950s, gamma scintillation detectors had been coupled

71
72 NUCLEAR MEDICINE: THE REQUISITES
to mechanical devices to produce scans of the thyroid
gland. These scans and uptake studies stimulated
the early development of the nuclear medicine field.
Therapy with radioiodine has been a primary objective
since the beginning of nuclear medicine. Thyroid diag-
nostic studies and therapy principles serve as the physio-
logic basis for much that we do today and hope to do in
the future.
Thyroid Anatomy and Physiology
The thyroid follicular cell synthesizes,stores,and secretes
thyroid hormones. An understanding of iodine metabo-
lism, thyroid physiology, and the diseases that result in
disordered pathophysiology of the thyroid is needed for
the optimal performance and proper interpretation of
thyroid uptake studies and thyroid scintigraphy.
Anatomy
The thyroid gland is located at the anterior superior
aspect of the trachea just below the thyroid cartilage and
extends laterally, superiorly and inferiorly (Fig. 5-1). The
name of the gland is derived from the Greek word for
Figure 5-1 Thyroid gland. Anatomical relationship of the
thyroid gland to the thyroid and cricoid cartilages and other
adjacent anatomical structures.
shield. Because of its embryological development from
pharyngeal pouches and descent, ectopic tissue can be
found anywhere from the foramen caecum at the base of
the tongue to the myocardium. The pyramidal lobe
extends towards the hyoid bone and is a remnant of the
thyroglossal duct.
The normal adult thyroid gland weighs approximately
15–20 g. The gland consists of many follicles of varying
size lined by epithelium made up of cuboidal and colum-
nar follicular cells, which secrete toward the large lumen
of the follicle containing colloid (Fig. 5-2). The thyroid
gland is 50–75% colloid by weight.
Physiology
Iodine is essential for synthesis of thyroid hormones.
After oral ingestion, iodine is rapidly reduced to iodide in
the upper small intestine. More than 90% of the iodide is
systemically absorbed within 60 minutes of oral inges-
tion. It distributes in the blood as an extracellular ion sim-
ilar to chloride. Most leaves the extracellular space
through thyroid extraction (20%) or urinary excretion
(80%). Some is taken up by the salivary glands and gastric
mucosa, which secrete into the gastrointestinal tract.
Iodide Trapping and Organification
The thyroid follicular cell traps iodide by means of a high-
energy sodium iodide “thyroid pump” that concentrates
iodine intracellularly at 25–500 times the plasma concen-
tration. Trapping can be blocked competitively by mono-
valent anions (e.g., perchlorate). In the normal thyroid,
organification promptly follows trapping (see Fig. 5-2).
The iodide is oxidized by thyroid peroxidase at the fol-
licular cell colloid interface to neutral iodine, which
binds to tyrosine residues on thyroglobulin. These
mono- and di-iodinated tyrosines (MIT, DIT) couple to
form T3 and T4, which are stored in the colloid filled fol-
licular lumen. Organification can be blocked by drugs
such as propylthiouracil and methimazole.
Thyroid Hormone Storage and Release
Thyroid stimulating hormone (TSH) initiates iodide
uptake and organification,as well as release of thyroid hor-
mone through hydrolysis of thyroglobulin. Thyroglobulin
does not normally enter the bloodstream except during
disease states (e.g.,thyroiditis or thyroid cancer). The nor-
mal thyroid gland contains a 1-month supply of hormone,
thus drugs blocking hormone synthesis (e.g., propyl-
thiouracil) do not become fully effective in controlling
hyperthyroidism until intrathyroidal stores are depleted.
Thyroid-Pituitary Feedback
The thyroid-pituitary feedback mechanism is very sensi-
tive to circulating serum thyroid hormone levels and is
the dominant method of adjusting TSH secretion
(Fig. 5-3). When serum thyroid hormone levels are
increased, the serum TSH is suppressed; when serum thy-
roid hormone levels are low, serum TSH increases. The
major hormone released by the thyroid is T4, which is
 Endocrine System 73

Figure 5-2 Iodine metabolism.The thyroid follicular cell epithelium extracts (traps) iodide from
the plasma via the thyroid pump and organifies it.The iodide (I−) is converted to neutral iodine (I0)
which is then incorporated into thyroglobulin-bound tyrosine molecules as mono or diiodotyrosine
(MIT, DIT). Coupling of the iodotyrosines results in T4 and T3 bound to the thyroglobulin which is
transported to and stored in the colloid until T4 and T3 are released into the plasma by proteolytic
enzymes.

transported to peripheral tissues by thyroid-binding pro-
teins and converted to the more metabolically active T3
at peripheral tissue site of action.
Radiopharmaceuticals
Radioiodine
Because radioiodine is selectively trapped and organified
by the thyroid and incorporated into thyroid hormone,
radioactive iodine is an ideal physiological radiotracer,
providing clinically important physiological information
regarding thyroid function. I-123 and I-131 are the two
radiopharmaceuticals used clinically.
Because of the rapid absorption, prompt uptake, and
organification of iodine, radioactivity is detectable in the
thyroid gland within minutes and normally reaches the
thyroid follicular lumen by 20–30 minutes. A progres-
sive increase in thyroid uptake normally occurs over 24
hours (Fig. 5-4). The time delay between radioiodine
ingestion and imaging (e.g., 2–6 hours for routine I-123
thyroid imaging and 1–3 days for I-131) is dictated more
by the desire for background clearance and a high target-
to-background ratio than by slow gland uptake.
Radioiodine is also taken up in the salivary glands,
stomach, and to a lesser extent, choroid plexus. It is not
concentrated in these organs. The kidneys and gastroin-
testinal tract serve as the excretory route (Fig. 5-5).
Iodine I-131
Physics The physical half-life of I-131 is 8 days. It
undergoes beta minus decay and emits a principle primary
gamma photon of 364 keV (Table 5-1). The 364-keV
gamma photons are not optimal for modern-day gamma
74 NUCLEAR MEDICINE: THE REQUISITES
Figure 5-3 Thyroid-pituitary feedback loop. The normal
thyroid is under the control of thyroid-stimulating hormone (TSH).
The hypothalamic production of thyroid-releasing hormone (TRH)
and the pituitary release of TSH are decreased or suppressed as
circulating levels of thyroid hormone increase.
cameras. Camera count detection sensitivity for I-131 is
poor; approximately half of the photons penetrate the
typical three-eighths-inch crystal and thus are not
detected. Septal penetration of the collimator by the
high-energy emissions results in image degradation. High-
energy beta particles are also emitted, the principle one
being 0.606 MeV.
Dosimetry The high-energy beta emissions and
long physical half-life of I-131 result in relatively high
radiation to the patient, particularly to the thyroid
(approximately 1 rad/μCi) (Table 5-2). This high radiation-
absorbed dose severely limits the dose that can be
administered,further impacting on image quality.
Iodine I-123
Physics I-123 decays by electron capture with
a half-life of 13.2 hours. The principal gamma emission
is a 159 keV photon which is well-suited for gamma
cameras. There are a small percentage of higher energy
Figure 5-4 Thyroid radioiodine uptake. Radioiodine uptake
normally increases progressively over 24 hours (gray area) and is
between 10 and 30% at 24 hours. A typical Graves’ hyperthyroid
patient is noted by the broken line above the normal range with
24-hour uptakes ranging from 50–80%. Some patients with Graves’
disease have rapid iodine turnover which is depicted by the top
curve, with early elevated uptake but mildly elevated or normal
uptake at 24 hours.
emissions (2.4% 440–625 keV) and 0.15% (625–784 keV).
There are no particulate emissions (seeTable 5-1).
Dosimetry Methods used for I-123 production in
the United States today result in long-lived radionuclide
impurities.In the past,I-123 was contaminated with I-124
and I-125. However, commercially available I-123
produced by MDS Nordian in Canada is 99.9% I-123. The
maximal levels of identified impurities are Te-121 (0.05%)
and I-125 (0.06%). The thyroid receives 1.5–2.6 rads
(15–25% RAIU) from a 200 μCi dose of I-123 (Table 5-2).
The considerably lower radiation dosimetry of I-123
compared to I-131 allows administration of 200–400 μCi
of I-123 for routine thyroid scanning compared to 50 μCi
of I-131. This higher administered dose results in
considerably better image quality.
Tc-99m Pertechnetate
Because of Tc-99m pertechnetate’s low cost and ready
availability from molybdenum-99/Tc-99m generator sys-
tems, it has long served as an alternative to radioiodine
for thyroid scintigraphy.
Physics
The 140-keV photopeak of Tc-99m is ideal for use with
the gamma camera. It has a short 6-hour half-life and no
particulate emissions (see Table 5-1).
 Endocrine System 75

Figure 5-5 Radioiodine distribution within the body. I-123 thyroid whole body scan of patient
post total thyroidectomy for thyroid cancer who has received radioactive iodine therapy in the past,
thus no thyroid is seen. Otherwise, the distribution at 24 hours is normal with salivary gland and
gastric uptake and urinary excretion.

Pharmacokinetics
In contrast to the oral administration of radioiodine,
Tc-99m pertechnetate is administered intravenously.
Tc-99m is trapped by the thyroid in an identical manner
as iodide, but it is not organified nor incorporated into
thyroid hormone. Because it is not organified, it is not
retained in the thyroid. Thus, thyroid imaging is per-
formed at peak uptake 20–30 minutes after injection.
Dosimetry
The lack of particulate emissions and the short half-life of
6 hours results in relatively low radiation dosimetry to
the thyroid (see Table 5-2). Thus, the allowable adminis-
tered activity of Tc-99m pertechnetate (3–5 mCi) is con-
siderably higher that I-123 for routine thyroid scans. The
large photon flux provides high quality images.
Which Radiopharmaceutical to Use
I-131 is not used for routine thyroid scans because of its
poor image quality and high radiation dosimetry. On the
other hand, its long half-life permits delayed imaging that
improves the target-to-background ratio, increasing
detectability of substernal goiter and thyroid cancer.
76 NUCLEAR MEDICINE: THE REQUISITES

thyroid. Increasing amounts of iodine in the normal diet

Table 5-1 Physical Characteristics of Thyroid over the years has resulted in a lower normal value for
Radiopharmaceuticals the %RAIU. Numerous noniodine-containing drugs also
affect thyroidal uptake (see Box 5-1).
Tc-99m Suppression of uptake by exogenous iodine may pre-
pertechnetate I-123 I-131
clude successful imaging or accurate uptake measure-
MODE OF DECAY Isometric Electron Beta minus ments. As little as 1 mg of stable iodine can cause
transition capture marked reduction of uptake and 10 mg can effectively
PHYSICAL HALF-LIFE 6 hr 13.2 hr 8.1 days block the gland (98% reduction). Radiographic contrast
(T ⁄ )
1
2
media are a common source of iodine that interferes with
PHOTON ENERGY 140 keV 159 keV 364 keV radioiodine thyroid studies. A food and drug history
should be obtained from all patients prior to undergoing
ABUNDANCE 89% 83.4% 81%
thyroid imaging and uptake studies.
BETA EMISSIONS 606 keV
Chronic renal failure impairs iodide clearance,
expands the iodide pool, and thus lowers the %RAIU.
Hypothyroidism slows clearance of radioactive iodine
from the body; hyperthyroidism increases the clearance
Table 5-2 Dosimetry of Thyroid rate.
Radiopharmaceuticals Pregnancy
The fetal thyroid concentrates radioiodine after 10–12
weeks of gestation. Radioiodine crosses the placenta and
Tc-99m
pertechnetate I-123 I-131 significant exposure of the fetal thyroid can occur after
(rads/5 mCi, (rads/200 mCi, (rads/50 mCi, therapeutic doses to the mother, resulting in fetal
cGY/185 MBq) cGy/7.5 MBq) cGy/3.7 MBq) hypothyroidism and even cretinism.
THYROID 0.600 1.5 to 2.6* 39.000 to
Nursing Mother
65.000* Radioiodine is excreted in human breast milk. Because of
BLADDER WALL 0.430 0.070 0.150
its long half-life, nursing should be discontinued after
diagnostic or therapeutic studies with I-131.With I-123, it
STOMACH 0.250 0.050 0.085
has generally been recommended that breastfeeding may
SMALL INTESTINE 0.550 0.030 0.003 resume after 2–3 days.With Tc-99m pertechnetate, nurs-
RED MARROW 0.100 0.060 0.007 ing can be resumed after 24 hours.
TESTIS 0.050 0.027 0.006 Patient Information
OVARIES 0.150 0.072 0.009 Thyroid studies must always be interpreted in light of the
TOTAL BODY 0.070 0.009 0.035
patient’s clinical history, thyroid physical examination,
and, importantly, with knowledge of the patient’s serum
*Lower estimate assumes a 15% RAIU and higher estimate assumes a 25% RAIU, thyroid function studies.
both at time of calibration.

Methodology for Thyroid Uptake Studies

However, for reasons discussed later, I-123 is replacing
I-131 even for these indications. The high-energy beta
emissions of I-131 make it very useful for radiotherapy
of Graves’ disease, toxic nodules, and thyroid cancer.
Tc-99m pertechnetate is sometimes preferred for thyroid
scanning in children because of its low radiation dosime-
try and high count rate.
Special Considerations and Precautions
Food and Medications Containing Iodine
Stable iodine contained in foods and medications can
interfere with radionuclide thyroid studies ( Box 5-1).
Expansion of the iodine pool due to ingestion or par-
enteral administration of iodine containing agents results
in a reduced percent radioiodine uptake (%RAIU) by the
and Thyroid Scans
Thyroid uptake studies and thyroid scans are often per-
formed at the same clinic visit. However, they are usually
acquired with different instrumentation, providing differ-
ent but complementary information. Thyroid scans are
acquired with a gamma camera. Thyroid uptake studies
are most commonly acquired with a nonimaging gamma
scintillation probe detector. Camera-based methods are
sometimes used and will be discussed. A thyroid uptake
provides quantitative information regarding the percent
of the administered activity taken up by the thyroid.
Thyroid Uptake
Radioiodine Percent Uptake
Both I-131 and I-123 can be used for calculation of the
%RAIU or the percent of the administered radioactive
 Endocrine System 77

Box 5-1 Drugs, Foods, Radiographic Contrast Agents, and Therapies that Decrease or
Increase the %RAIU

DECREASED UPTAKE DURATION OF EFFECT

Thyroid Hormones
Thyroxine (T4) 4–6 weeks
Triiodothyronine (T3) 2 weeks

Excess Iodine (Expanded Iodine Pool)

Saturated solution of potassium iodide 2–4 weeks
Some mineral supplements, cough medicines, and vitamin preparations 2–4 weeks
Iodine food supplements
Iodinated drugs (e.g., amiodarone) Weeks to months
Iodinated skin ointments 2–4 weeks
Congestive heart failure
Renal failure

Radiographic Contrast Media

Water-soluble intravascular media 2–4 weeks
Oral cholecystographic agents 4 weeks to indefinite
Fat-soluble media (lymphography) Months to years

Noniodine-Containing Drugs Variable

Adrenocorticotropic hormone, adrenal steroids
Monovalent anions (perchlorate)
Penicillin
Antithyroid drugs
Propylthiouracil (PTU) 3–5 days
Methimazole (Tapazole) 5–7 days
Bromides

Goitrogenic foods (e.g., cabbage, turnips)

Prior radiation to neck

INCREASED UPTAKE
Iodine Deficiency
Pregnancy
Rebound after therapy withdrawal (thyroid hormones, antithyroid drugs)
Lithium

iodine taken up by the thyroid. Clinical indications for
uptake determinations are few, but important (Box 5-2).
Indications The most common clinical indication
for a %RAIU study is to aid in the differential diagnosis
of newly diagnosed thyrotoxicosis. In most cases, the
referring physician seeks to differentiate Graves’ disease,
the most common cause for thyrotoxicosis (Box 5-3),from
other causes (e.g., thyroiditis, the second most common
cause). Therapy of Graves’ disease is quite different than
that for other causes. The %RAIU is elevated in Graves’
disease, but suppressed or decreased in most other causes
of thyrotoxicosis with diffuse goiter, such as thyroiditis
(Box 5-4).
The %RAIU is also often used for the calculation of an
I-131 therapy dose for Graves’ disease (Box 5-5).
Methodology Medications that might interfere
with thyroid uptake should be discontinued for an
appropriate length of time (see Box 5-1). Patients should
have nothing by mouth for 4 hours prior to the study to
assure good radioiodine absorption. I-123 and I-131 are
usually administered in capsule form rather than liquid.
The unit-dosed capsule formulation minimizes airborne
exposure of radioiodine to technologists and is convenient
for handling.
When a scan is not needed, 5–10 μCi I-131 or 50–100
μCi I-123 is adequate for an uptake because of the
probe’s high detection sensitivity compared to a gamma
camera.When a scan is indicated, both can be performed
using the scan dose of I-123 (200–300 μCi). The stan-
dard %RAIU uptake is acquired at 24 hours. In some
78 NUCLEAR MEDICINE: THE REQUISITES

BOX 5-2 Clinical Indications for Thyroid BOX 5-5 Calculation of Thyroid Percent
Scintigraphy and Uptakes Uptake of Radioiodine I-123 and
I-131
THYROID SCANS THYROID UPTAKES
1. PRELIMINARY MEASUREMENTS
Determination of functional Differential diagnosis of
status (cold, hot) of thyrotoxicosis Place dose capsule in neck phantom and count for
thyroid nodule 1 minute
Detection of ectopic Estimate I-131 therapy Count patient’s neck and thigh (background) for
thyroid tissue (lingual dose for Graves’ 1 minute
thyroid) disease
Differential diagnosis of Together with whole 2. ADMINISTER ORAL DOSE CAPSULE
mediastinal masses body thyroid cancer
(substernal goiter) scans 3. UPTAKE MEASUREMENT AT (OPTIONAL: 2–6 HOURS
AND) 24 HOURS
Thyroid cancer whole Estimate residual
body scan thyroid postsurgery Count patient’s neck for 1 min
Estimate I-131 therapeutic Count patient’s thigh for 1 min
effectiveness
4. CALCULATION AT SELECTED UPTAKE INTERVALS AT
Follow-up for (OPTIONAL: 2–6 HOURS AND) 24 HOURS
recurrence
Neck (background corrected) counts/min
%RAIU = × 100
Dose capsule (decay corrected and
background corrected) counts/min
BOX 5-3 Clinical Frequency of Various
Correction is also routinely made for room background. In the past, a standard
Causes for Thyrotoxicosis dose capsule was also counted initially and at uptake intervals and used for decay
correction. However, in most modern systems, decay is automatically calculated
by the probe computer system.
Graves’ disease 70%
Thyroiditis 20%
Toxic multinodular goiter 5%
Toxic adenoma 5%
Others <1% clinics, a 2–6 hour uptake is also routinely performed.
In some clinics, only a 2–6 hour uptake is done.
A nonimaging gamma scintillation probe detector is
used for radioiodine thyroid uptake studies. It has a 2-cm
Box 5-4 Differential Diagnosis thick × 2-cm diameter sodium iodine crystal with an
of Thyrotoxicosis-based open cone-shaped single-hole lead collimator coupled to
Increased or Decreased %RAIU a photomultiplier tube and electronics.
Room background activity is determined. The
radioiodine capsule is placed in a Lucite neck phantom
INCREASED UPTAKE
and activity counted with the probe detector placed at
Graves’ disease
a standardized distance of 30 cm (Fig. 5-6). The capsule
Multinodular toxic goiter
is administered to the patient. The probe is placed 30
Hashitoxicosis
Central hypothyroidism cm from the anterior surface of the patient’s neck, such
Hydatiform mole, trophoblastic tumors, that the entire gland can be detected by the probe but
choriocarcinoma most extrathyroidal activity is not. The patient’s neck
Metastatic thyroid cancer (background) is counted (Fig. 5-7).
At the uptake times (2–6 hours and/or 24 hours),
DECREASED UPTAKE counts are obtained for the neck and the patient’s thigh
Subacute thyroiditis (background). The percent radioiodine uptake is calcu-
Granulomatous thyroiditis (de Quervain’s) lated according to this formula:
Silent thyroiditis
Postpartum thyroiditis Neck counts/min (background corrected)
%RAIU = × 100
Iodine-induced thyrotoxicosis (Jod-Basedow) Administered dose capsule counts/min
Amiodarone-induced thyrotoxicosis (background and decay corrected)
Thyrotoxicosis factitia
Struma ovarii In the past, a standard reference capsule similar in activ-
ity to the administered capsule was counted initially and

Figure 5-6 Uptake probe: neck lucite phantom. The gamma
detector is placed at a standard distance of 30 cm from the neck
phantom which contains the capsule dose of radioiodine to be
administered to the patient.
at the uptake intervals. The purpose of the standard was
to correct for decay. A dose-to-standard ratio was deter-
mined to correct for the difference in standard and
administered dose. In present-day uptake probe-computer
systems,decay is automatically corrected.
Normal Values The %RAIU increases progressively
over 24 hours (see Fig 5-4). In the United States, the
normal range is approximately 10–30% (8-35% in some
laboratories) at 24 hours. Before widespread iodine
supplementation in bread and table salt, the normal
range was substantially higher. The normal range for the
2–6 hour %RAIU is 4–15%.
The early 2–6 hour %RAIU can serve two purposes.
An occasional patient with Graves’ disease has rapid
iodine turnover within the thyroid, resulting in a normal
or only mildly elevated 24 hour %RAIU but a very high
2–6 hour value (see Fig 5-4). Thus an early uptake con-
firms that the %RAIU is indeed elevated, and furthermore
may suggest that a higher therapy dose is needed
Endocrine System 79
Figure 5-7 Uptake probe: patient. The detector is positioned
30 cm from the patient’s thyroid.
because of the rapid turnover and shorter residence time
in the thyroid.
In some clinics, the 2–6 hour %RAIU is used to predict
the 24-hour uptake so that a therapy dose can be esti-
mated, ordered, and available for administration at
24 hours. The dose can be adjusted based on the 24 hour
uptake. Other clinics treat immediately after the 2–6 hour
%RAIU.
Another use of the %RAIU is in conjunction with
whole body thyroid cancer scans (i.e., to quantify resid-
ual thyroid tissue uptake after thyroidectomy or I-131
therapy, therapeutic effectiveness, or recurrence). For
whole body thyroid cancer scans, a camera-based
method is used to calculate the %RAIU. A standard with
calibrator-measured activity is also imaged. Regions of
interest are drawn for the thyroid, background, and stan-
dard. The uptake is calculated at the time of whole body
imaging (24–48 hours).
Tc-99m Pertechnetate Uptake
The advantages of using Tc-99m pertechnetate as an
alternative to radioiodine to calculate an uptake include
its favorable radiation dosimetry, a particular considera-
tion for children (see Table 5-2), and that the study can be
80 NUCLEAR MEDICINE: THE REQUISITES

completed within 30 minutes with uptake results avail-
able soon thereafter. The disadvantages of the Tc-99m
pertechnetate uptake include a less well-defined normal
range, the standard practice of calculating the therapy
dose based on the radioiodine uptake, and the lack of
software for this calculation on some newer camera com-
puter systems.
Methodology A scintillation probe is not used for
a Tc-99m uptake because of the high neck and body
background. This is a gamma camera technique for
calculating the percent uptake, similar to that used for
thyroid cancer scans. Before and after injection of the
Tc-99m pertechnetate, the syringe is imaged with the
gamma camera (preinjection counts minus postinjection
residual counts = administered counts). Twenty minutes
after injection, the thyroid scan is acquired on computer.
Regions of interest are drawn on computer for the
thyroid, thyroid background, and the syringes. Areas of
interest are normalized for pixel size and thyroid and
syringe counts are normalized for time of acquisition.
The percent uptake is calculated:
Thyroid counts –
Tc-99m pertechnetate Background counts
= × 100
percent uptake Injected counts – R counts
Normal Tc-99m uptake ranges from 0.3–4.5%. Accuracy
is less than with the %RAIU.
A simple qualitative approach has been used to esti-
mate uptake by obtaining images with the salivary glands
in the same field-of-view as the thyroid. In the euthyroid
patient, relative uptake in the salivary and thyroid glands
is similar.With hyperthyroidism, thyroid uptake is consid-
erably greater than salivary gland uptake.
Procedure Radioiodine is administered orally. The
usual I-123 thyroid scintigraphy dose is 200–300 μCi.
The scan is usually acquired at 2–6 hours after administra-
tion but may be acquired at the time of the 24-hour
%RAIU. The higher count rate obtainable at 2–6 hours
allows for shorter imaging time and better image quality.
The low count rate at 24 hours requires longer acquisition
time which increases the likelihood of patient movement.
With Tc-99m pertechnetate, 3–5 mCi is administered intra-
venously and imaging begins 20–30 minutes after injection.
For both radiopharmaceuticals, a standard or large
field-of-view gamma camera is used, equipped with a pin-
hole collimator (Fig. 5-8) that has an interchangeable lead
pinhole insert of 3- to 6-mm in internal diameter placed
in its distal aspect. Smaller diameter inserts provide
higher resolution but lower sensitivity.
A 15–20% photopeak window is set at 159 keV for
I-123 and at 140 keV for Tc-99m. Imaging protocols for
thyroid imaging for the two radiopharmaceuticals are
similar and described in more detail in Boxes 5-6 and 5-7.
The patient is positioned supine with the neck hyperex-
tended so that the plane of the thyroid gland is parallel to
the crystal face of the camera. The thyroid gland should
fill approximately two-thirds of the field of view. This
is achieved with a 6–8 cm distance from the collimator

Thyroid Scintigraphy
The thyroid scan depicts the entire gland in a single image
and permits direct correlation of physical findings with
abnormalities in the image. The combination of gamma
camera and pinhole collimator offers the flexibility of
obtaining multiple-view high-resolution images of the thy-
roid. Pinhole collimator magnification provides image res-
olution superior to parallel-hole collimators in the range of
5 mm,compared to 1–2 cm with a parallel-hole collimator.
Thyroid Examination
The thyroid gland should be routinely examined by pal-
pation at the time of imaging in order to estimate the size
of the gland and to confirm the presence and location of
nodules. A radioactive marker source (122-keV Cobalt-
57 or Tc-99m) can then be placed over the palpated nod-
ule for anatomical and functional correlation.
Methodology
Thyroid I-123 and Tc-99 Pertechnetate Scans
Figure 5-8 Pinhole collimator. The gamma camera with
Clinical indications for thyroid scintigraphy are listed in a pinhole collimator is placed close to the thyroid to permit
Box 5-2 and discussed later in the section entitled maximal magnification. The thyroid gland should fill
Clinical Indications for Thyroid Scintigraphy. approximately two thirds of the field of view.

BOX 5-6 Tc-99m Pertechnetate Thyroid
Imaging: Protocol Summary
PATIENT PREPARATION
Discontinue any medications that interfere with thyroid
uptake of Tc-99m pertechnetate.
Nothing by mouth for 4 hours prior to study.
RADIOPHARMACEUTICAL
Tc-99m pertechnetate, 3–5 mCi (111–185 MBq)
intravenously
TIME OF IMAGING
20 min after radiopharmaceutical administration
IMAGING PROCEDURE
Gamma camera with a 3- to 6-mm aperture pinhole
collimator and a 20% energy window centered at
140 keV.
Position the patient supine with the chin up and neck
extended.
Position the collimator so that the thyroid fills about two
thirds of the diameter of the field of view.
Obtain anterior, 45-degree LAO and RAO views (move
the collimator rather than the patient).
Obtain 250k counts per view.
Mark the chin and suprasternal notch.
Note the position and mark palpable nodules and
surgical scars.
Place marker sources lateral to the thyroid to calibrate size.
to the surface of the neck. Magnification increases as the
pinhole collimator approaches the neck.
On one image, a radioactive marker (Tc-99m or
Cobalt-57) or computer cursor is placed at the sternal
notch and on the right side (Figs. 5-9 and 5-10). A 4- to
5-cm line source marker or two point sources 4–5 cm
apart may be placed on the neck just lateral to the thy-
roid lobes and parallel to their long axis (Fig. 5-11) to esti-
mate the size of the thyroid and nodules. Because of the
three-dimensional nature of the gland and pinhole colli-
mator distortion, this is an approximate measurement
and does not obviate physical exam size estimation.
Images are routinely obtained in the anterior, right
anterior oblique (RAO), and left anterior oblique (LAO)
views (see Fig. 5-9). Each image is obtained for 100,000
counts. It is preferable for the patient to remain in one
position while the camera and collimator are moved to
the different projections, thus making images more
reproducible between patients and resulting in less
image distortion and patient motion.
Additional images with a radioactive marker may be
indicated to determine whether a palpable nodule takes
Endocrine System 81
BOX 5-7 Iodine-123 Thyroid Imaging:
Protocol Summary
PATIENT PREPARATION
Discontinue any medications that interfere with thyroid
uptake of radioiodine.
Nothing by mouth for 4 hours prior to study.
RADIOPHARMACEUTICAL
I-123, 100–400 μCi (3.7–15 MBq), orally in capsule form
TIME OF IMAGING
At 4–6 or 24 hours
IMAGING PROCEDURE
Use a gamma camera with a 3- to 6-mm aperture pinhole
collimator and a 20% energy window centered at 159
keV.
Position the patient supine with the chin up and the
neck extended.
Position the collimator so that the thyroid fills about
two-thirds of the diameter of the field of view.
Obtain anterior, 45-degree LAO and RAO views (move
the collimator, if possible, rather than the patient).
Obtain 100k–250k counts per view.
Mark the chin and suprasternal notch.
Note the position and mark palpable nodules and
surgical scars.
Place marker sources lateral to the thyroid to calibrate
size.
up the radiopharmaceutical (i.e., a hot or cold nodule).
Special care should be taken to avoid the parallax effect of
a pinhole collimator. The parallax effect results in
a change in the relationship between a near and distant
object when viewed from different angles.With a pinhole
collimator, this can result in misregistration of the relation-
ship between the nodule and the marker or, in the case of
a suspected substernal goiter, the suprasternal notch
marker and the thyroid. The effect can be minimized in
several ways. One method is to obtain an image with the
collimator at an increased distance from the thyroid (see
Fig. 5-9), decreasing the effect of magnification and distor-
tion. A second method is to place the marker region of
interest in the center of the field-of-view. Finally, a parallel-
hole collimator might be used for the marker image.
Iodide I-131 Scan Because of the high radiation
dose the thyroid with I-131, clinical indications for I-131
thyroid scans are limited to confirming the thyroid origin
of a mediastinal mass (substernal goiter) and for thyroid
cancer scintigraphy. The advantage of I-131 is that delayed
imaging allows for improved target-to-background ratio,
thus improving detectability.
82 NUCLEAR MEDICINE: THE REQUISITES
Figure 5-9 Normal I-123 thyroid scan. On the initial image, the
collimator is placed at a greater distance from the neck than the
other images. A computer cursor marks the suprasternal notch
(SSN) and the right side (RT).The collimator is moved closer to the
neck to acquire the anterior, right anterior oblique and left anterior
oblique views, which have greater magnification and resolution.
For the substernal goiter, 50 μCi I-131 is administered
orally. A large-field-of-view gamma camera equipped with
a high-energy parallel-hole collimator is used. A 20–30%
window is centered at 364 keV. A radioactive marker
should be placed at the suprasternal notch as discussed
previously (see Fig.5-11)
Thyroid Cancer Scan Thyroid cancer cells are
hypofunctional compared to normal thyroid tissue and thus
take up radioiodine to a lesser degree. This is the reason
that cancer nodules appear cold on routine thyroid scans.
To maximize thyroid cancer cell uptake,TSH stimula-
tion is necessary. For decades, this has been done through
endogenous stimulation of TSH by making the patient
hypothyroid by total thyroidectomy and radioactive iodine
therapy. Endogenous stimulation of TSH requires discon-
tinuation of thyroid hormone (i.e., 4–6 weeks off
Synthroid and 2 weeks without Cytomel). The serum TSH
should be greater than 30 IU before radioiodine can be
administered. Alternatively, recombinant TSH (Thyrogen)
can stimulate uptake without thyroid hormone with-
drawal. A low-iodine diet is an ancillary method used to
increase uptake by decreasing the iodine pool.
Iodine-131 For many years, 5–10 mCi of I-131 was
the usual administered dose for whole-body thyroid can-
cer scans. However, because of reports of thyroid “stun-
ning” after the diagnostic dose, the recommended dose
was reduced to 2 mCi.With diagnostic doses greater than
5 mCi, this stunning of thyroid cells decreases subse-
quent therapeutic dose uptake.
When Thyrogen (recombinant TSH) is used as an alter-
native to endogenous TSH stimulation, the dose is raised
to 4 mCi because uptake is 50% of that seen with thyroid
hormone withdrawal. The explanation is that thyroid
hormone withdrawal and the resulting hypothyroidism
causes renal insufficiency with reduced radioiodine
clearance and increased iodine uptake. Thyrogen doses
do not have that advantage.
Thyrogen is given on two consecutive days as an intra-
muscular injection of 0.9 mg. On the third day, radioio-
dine is administered. Imaging is performed on day 5 for
I-131 and day 4 for I-123. Serum thyroglobulin, a tumor
marker for thyroid cancer, is drawn on day 5.
A large field-of-view gamma camera equipped with
a high-energy parallel-hole collimator is used with
a 20–30% window centered at 364 keV. Imaging is usu-
ally performed 48 hours after oral administration. For
detection of thyroid carcinoma, the most important view
is the anterior image of the head, neck, and chest.
However,whole body views are desirable because thyroid
metastases may occur outside this field of view (e.g., in
the bones of the skull, humeri, and femurs, liver, and
brain). Spot imaging requires 10–20 minutes per view.
The pinhole collimator allows for high-resolution views
of the neck, which is most helpful when there is intensive
uptake in the thyroid bed and associated “star” artifact
(Fig. 5-12) due to septal penetration of the parallel-hole
collimator. Since pinhole collimators do not have septa,
neck uptake can be resolved without this artifact.
Iodine-123 I-123 is increasingly replacing I-131 for
whole body thyroid cancer scans.Stunning is not an issue
with I-123, image quality is better, and the study is com-
pleted at 24 hours. One might expect I-123 to detect
fewer tumors than I-131 because of the earlier imaging
period (24 rather than 48 hours). However, investigations
have shown similar ability to detect metastases, possibly
because of the higher photon flux with I-123. The orally
administered dose of I-123 is typically 1.5–2 mCi.
Clinical Indications for Thyroid Uptake
Studies
Thyrotoxicosis
Thyrotoxicosis is characterized by hypermetabolism due to
increased circulating thyroid hormone. Hyperthyroidism
is thyrotoxicosis caused by a hyperfunctioning thyroid
gland (e.g., Graves’ disease or toxic nodular goiter)
(Box 5-8). Examples of thyrotoxicosis not caused by
a hyperfunctioning thyroid gland are subacute thyroiditis,
thyroiditis factitia,and struma ovarii.
Clinical Diagnosis
The symptoms of thyrotoxicosis are those of increased
metabolism such as weight loss, tachycardia, palpitations,
 Endocrine System 83

Figure 5-10 Substernal goiter. A, Chest radiograph reveals a superior mediastinal mass. B,
Computed tomography confirms the presence of the mass, which demonstrates inhomogeneous
density. C, Subsequent I-131 scintigraphy reveals a large substernal goiter. A radioactive marker is
placed at the suprasternal notch.

heat intolerance, hyperhidrosis, and anxiety. However,
these symptoms are nonspecific and require confirmation
with serum thyroid function studies. A suppressed
serum thyroid stimulating hormone (TSH) less than
0.1 mU/L is the most sensitive test for diagnosis of thyro-
toxicosis. The suppressed TSH is caused by negative feed-
back on the pituitary from the elevated serum thyroid
hormone. The only exception to a suppressed TSH with
thyrotoxicosis is a rare hypothalamic or pituitary etiology.
Differential Diagnosis
The clinical history and physical exam may suggest the
etiology of thyrotoxicosis. For example, recent upper
respiratory infection and a tender thyroid would suggest
subacute thyroiditis, whereas exophthalmus and pretib-
ial edema is consistent with Graves’ disease. However, in
many cases, the diagnosis is uncertain. The %RAIU can
help with this differential diagnosis.
%RAIU Interpretation
The %RAIU does not define thyroid function per se. An
increased %RAIU is not diagnostic of hyperthyroidism.
Patients with iodine deficiency,dyshormonogenesis (organ-
ification defects), or chronic autoimmune thyroiditis may
have elevated %RAIU despite clinical euthyroidism or even
hypothyroidism (Box 5-9). The %RAIU must be interpreted
in light of the patient’s serum thyroid function studies.
Normally, iodine uptake is dependent on pituitary-
produced TSH stimulation. For many causes of thyrotoxi-
cosis, radioiodine uptake will be suppressed (less than
1–2%) (e.g., in subacute thyroiditis, iodine-induced thy-
roiditis,and thyrotoxicosis factitia [see Box 5-8]).However,
iodine uptake will be increased despite the suppressed
TSH if the thyroid has autonomous function independent
of pituitary feedback, the most common cause being
Graves’ disease. Thus, the clinical utility of the %RAIU for
thyrotoxicosis lies in the confirmation or exclusion of
Graves’disease as the cause of a toxic diffuse goiter.
The %RAIU has been used in conjunction with various
pharmacologic interventions described in the later
section Other Thyroid Function Studies. These include
the T3 suppression test,TSH stimulation test,and perchlo-
rate washout test. Although these interventions nicely
delineate underlying pathophysiological processes, they
are rarely requested in current practice due to various
advancements in diagnosis and therapy.
84 NUCLEAR MEDICINE: THE REQUISITES
Figure 5-11 Solitary cold nodule. A palpable nodule
corresponds to the cold defect in left lower lobe on the thyroid
scan. Radioactive markers are placed 4 cm apart on the left side as
an aid to approximate the size of gland.
Diseases with Increased %RAIU
The list of diseases producing thyrotoxicosis with an “ele-
vated” uptake (see Box 5-4) includes entities where the
uptake may be in the normal range (e.g., toxic nodules).
This can be viewed as elevated in the sense that the nor-
mal response to thyrotoxicosis is suppression of TSH by
T-4 feedback to the pituitary with subsequent suppres-
sion of iodine uptake and thus %RAIU.
Graves’ Disease Between 70% and 80% of patients
presenting with thyrotoxicosis have Graves’ disease as the
etiology. This autoimmune disease is most commonly seen
in middle-aged females but also occurs in children and the
elderly. Thyroid-stimulating immunoglobulins similar to
TSH bind to the follicular cells causing hyperplasia and
autonomous thyroid hyperfunction.
The diagnosis of Graves’ disease is often straightfor-
ward, such as thyrotoxicosis, diffuse goiter without nod-
ules, characteristic exophthalmopathy, and pretibial
dermopathy. However, in many patients the diagnosis is
suspected but uncertain. An elevated %RAIU, usually in
the range of 50–80%,confirms the diagnosis of Graves’dis-
ease and excludes most other causes. A Tc-99m pertech-
netate uptake in Graves’ disease is typically greater than
4.0%. A thyroid scan is often not necessary to confirm
Graves’ disease if physical examination or sonography is
consistent with diffuse goiter without nodules.
Multinodular Toxic Goiter Sometimes referred to
as Plummer’s disease, multinodular goiter often presents
in older patients with tachyarrhythmias, weight loss,
depression, anxiety, and insomnia. The hypermetabolism
may exacerbate other medical problems. The natural
history of individual autonomously functioning thyroid
follicular adenomas is hemorrhage and necrosis,
ultimately becoming nonfunctioning nodules. However,
the time course is prolonged and variable, and the disease
requires prompt therapy. Physical examination of the
thyroid or ultrasonography can confirm nodularity but
not function. Serum thyroid function tests make the
diagnosis of thyrotoxicosis not infrequently T-3 toxicosis.
The %RAIU is often only moderately elevated; it can
even be in the high normal range. The thyroid scan has
the classical picture of high uptake within hyperfunc-
tioning nodules but suppression of the extranodular tis-
sue due to pituitary feedback and TSH suppression.
Therapy with radioactive iodine I-131 is the usual
treatment.
Single Autonomously Functioning Thyroid Nodule
Toxic nodules occur in only 5% of patients with a solitary
palpable nodule. Most are nonfunctioning. This entity
occurs more frequently in the elderly and those living in
iodine deficient regions. Once an autonomous nodule
grows to a size of 2.5–3.0 cm, it produces enough thyroid
hormone to cause clinical thyrotoxicosis. Therapy is
often with radioactive iodine, although surgery is a treat-
ment choice. The %RAIU is usually in the normal range
and the thyroid scan shows uptake in the nodule but sup-
pression of the remainder of the gland.
Chronic Thyroiditis (Hashitoxicosis) Hashimoto’s
disease commonly presents in middle-aged females as
goiter or hypothyroidism. Much less frequently, patients
manifest initially or later in the course of the disease with
thyrotoxicosis (3–5% of cases), so-called Hashitoxicosis.
Histopathologically, lymphocytic infiltration is predomi-
nant. Some consider Hashitoxicosis to be an overlap
syndrome with Graves’disease.
The gland is diffusely and symmetrically enlarged. It is
nontender, firm, and usually without nodules. Serum
antithyroglobulin and antimicrosomal antibodies are ele-
vated. During the thyrotoxic phase, the %RAIU is
increased, similar in degree to Graves’ disease. The thy-
roid scan typically shows diffuse increased uptake.
Radioactive iodine I-131 is the usual therapy.
Rare Causes A pituitary adenoma secreting TSH is
a rare cause of hyperthyroidism. An even rarer condition
is resistance of the pituitary to thyroid hormone
feedback. In both cases, the TSH is elevated. Hydatiform
mole, trophoblastic tumors, and choriocarcinoma may
manifest symptoms of hyperthyroidism.Human chorionic
gonadotropin (HCG) is a weak TSH-like agonist. The
serum level of HCG correlates with the severity of
hyperthyroidism.Serum TSH is suppressed and the %RAIU
is elevated. Hyperthyroidism secondary to metastatic
thyroid cancer is quite rare, usually occurring with
follicular carcinoma.

Diseases with Suppressed %RAIU
Subacute Thyroiditis The most common reason
for thyrotoxicosis associated with a decreased %RAIU is
subacute thyroiditis. There are several causes for the
disease. Granulomatous thyroiditis (de Quervain’s) is
characteristically preceded by several days of upper
respiratory illness and presents with a tender thyroid.
Histopathologically, granulomas are seen on biopsy.
Silent thyroiditis, commonly seen in elderly patients, is
not a granulomatous process nor is it associated with
respiratory symptoms or thyroid tenderness. Often,
elderly patients present with arrhythmia and normal size
Endocrine System 85
Figure 5-12 Star artifact.
Whole body thyroid cancer scan
obtained seven days after high-
dose I-131 therapy. The intense
uptake in the thyroid bed results
in a star artifact caused by septal
penetration of the high energy
photons through the septa of the
collimator. Note liver uptake
caused by metabolized thyroid
hormone.
thyroid. Postpartum thyroiditis occurs within weeks
or months of delivery. A mild goiter is often palpated
in patients with subacute thyroiditis.
The decreased %RAIU associated with subacute thy-
roiditis is the result of an intact pituitary feedback mech-
anism, not because of damage and dysfunction of the
gland (Fig. 5-13). Uptake is suppressed in the entire
gland, but the disease is often patchy or regional.
During the initial stage of subacute thyroiditis, stored
intracellular thyroid hormone is released into the blood.
This is caused by increased cell permeability as a result of
the inflammatory process. As the inflammation resolves,
86 NUCLEAR MEDICINE: THE REQUISITES

Box 5-8 Classification of Thyrotoxicosis

Based on Thyroid Hyperfunction
or No Thyroid Hyperfunction

Thyroid hyperfunction
A. Abnormal thyroid stimulator
1. Graves’ disease
2. Trophoblastic tumor
a. Hydatiform mole, choriocarcinoma uterus or
testes
B. Intrinsic thyroid autonomy
1. Hyperfunctioning adenoma
2. Toxic multinodular goiter
C. Excess production of TSH (rare)

No thyroid hyperfunction
A. Disorders of hormone storage
1. Subacute thyroiditis
2. Chronic thyroiditis with transient thyrotoxicosis
B. Extrathyroid source of hormone
1. Thyrotoxicosis factitia Figure 5-13 Subacute thyroiditis. Suppressed Tc-99m thyroid
2. “Hamburger toxicosis”(epidemic caused by uptake in patient presenting with new onset thyrotoxicosis.
thyroid gland contaminated hamburger meat)
3. Ectopic thyroid tissue
a. Struma ovarii
b. Functioning follicular carcinoma
ability to respond to TSH stimulation. The hypothy-
roidism usually resolves over weeks and months and the
serum thyroid hormone levels decrease and often fall TSH and %RAIU return to normal. The rapidity of this
into the subnormal range with a resulting rise in TSH. It is process depends on the degree of damage. The %RAIU
not uncommon for patients to become hypothyroid, depends on the stage of the disease (Fig. 5-14).
during the recovery phase, manifested by a high TSH. Iodine-Induced Thyrotoxicosis ( Jod-Basedow)
The level of the %RAIU depends on the damaged thyroid’s In the past,this condition occurred with the introduction

Box 5-9 Relationship of Thyroid Uptake to Thyroid Function

Thyroid Uptake

THYROID FUNCTION INCREASED NORMAL DECREASED

Thyrotoxicosis Graves’ disease Antithyroid drugs Expanded iodide pool
Hashitoxicosis Propylthiouracil Subacute thyroiditis, thyrotoxic
Methimazole phase
Thyrotoxicosis factitia
Antithyroid drugs
Struma ovarii
Euthyroid Rebound after antithyroid Decompensated
drug withdrawal dyshormonogenesis
Recovery from subacute thyroiditis
Compensated dyshormonogenesis
Hypothyroid Decompensated dyshormonogenesis Hashimoto’s disease Hypothyroidism:
Hashimoto’s disease after I-131 therapy primary or secondary
Subacute thyroiditis,
recovery phase
decompensated
dyshormonogenesis
 Endocrine System 87

Figure 5-14 Clinical course of subacute thyroiditis. Typical evolving pattern of the serum T4,TSH
and %RAIU over 9 months, from initial presentation to resolution. When the patient initially
presents, the T4 is elevated and TSH and RAIU are suppressed. Once there is no more thyroid
hormone to release, thyroid function may be poor due to inflammatory damage to the gland and the
TSH and RAIU will rise. With time, most patients become euthyroid and the values normalize.

of iodized salt into the diet in iodine deficient areas
(goiter belts). Today it most commonly occurs in
patients who have received iodinated contrast media
during a computed tomography (CT) exam. The iodine
induces a thyroiditis. The %RAIU is usually near zero.
Occasionally, the iodine load causes activation of
subclinical Graves’ disease or toxic multinodular goiter
and the uptake is elevated rather than suppressed.
Amiodarone-Induced Thyrotoxicosis Amiodarone
is an antiarrhythmic drug containing 75 mg iodine per
tablet. It has a physiological half-life of more than 3 months
and its effects may last even longer. Hyperthyroidism or
hypothyroidism occurs in up to 10% of patients on the
drug. Two types of thyrotoxicosis are seen: type I, which is
iodine-induced (Jod-Basedow) in patients with preexisting
nodular goiter or subclinical Graves’ disease, and type II,
which is more common and results in a destructive
thyroiditis. In the latter, the %RAIU is near zero. In the
former,the uptake may be elevated.
Thyrotoxicosis Factitia This is not a rare cause of
clinical hyperthyroidism. The thyroid hormone may
have been prescribed by a physician or surreptitiously
taken by the patient for weight loss. Oftentimes the
patients are healthcare workers.
Struma Ovarii Approximately 1–2% of benign ovarian
teratomas have functioning thyroid tissue as a major
component. In rare instances, this tumor produces
sufficient thyroid hormone to cause thyrotoxicosis. The
diagnosis is suspected in a patient with a concomitant
pelvic mass. The normal cervical thyroid is suppressed.
The ectopic thyroid tissue can be visualized with
radioiodine or Tc-99m pertechnetate.
I-131 Therapy Dose Calculation for Graves’ Disease
This important clinical indication for a %RAIU is dis-
cussed in the Radioiodine Therapy section later.
Radioiodine Uptake on Thyroid Cancer Scans
The %RAIU is frequently calculated at the time of whole
body thyroid cancer imaging. It can provide an estimate
of postthyroidectomy residual thyroid prior to I-131 abla-
tion therapy and on follow-up scans to quantify the effec-
tiveness of I-131 therapy. The amount of uptake seen
visually on the whole body thyroid scan can be mislead-
ing because of the variability of background clearance
and the windowing used for “optimal”viewing.
Clinical Indications for Thyroid
Scintigraphy
Thyroid scans have been used diagnostically for decades
for the evaluation of various types of thyroid disease.
They are requested less today than in the past because of
the availability of other thyroid imaging modalities and the
aggressive use of diagnostic percutaneous aspiration
biopsy of thyroid nodules. Because of the thyroid scan’s
88 NUCLEAR MEDICINE: THE REQUISITES
functional nature, scintigraphy still provides valuable clini-
cal information for many patients.
Normal Thyroid Scintigraphy
Thyroid scans should always be correlated with physical
examination of the thyroid gland and interpreted with
knowledge of the patient’s thyroid function studies and
other imaging studies.
The normal thyroid has a butterfly shape with lateral
lobes extending along each side of the thyroid cartilage
(see Fig. 5-9). The lateral lobes are connected by an isth-
mus that crosses the trachea anteriorly below the level of
the cricoid cartilage. However, the appearance of the
gland is quite variable from patient to patient. The right
lobe is often larger than the left. The lateral lobes meas-
ure 4–5 cm from superior to inferior poles and are 1.5–2
cm wide. The pyramidal lobe ascends from the isthmus
or adjacent part of either lobe (more often the left lobe)
to the hyoid bone.
The normal gland has homogeneous and uniform dis-
tribution of radiotracer throughout. Some increased
intensity may be seen in the middle or medial aspects of
the lateral lobes, owing to the thickness of the gland in
this location. The amount of activity in the isthmus
varies greatly, with little or no activity in some patients
and prominent activity in others. In normal adults, the
thin pyramidal lobe is usually not seen.
The salivary glands are routinely seen on Tc-99m
pertechnetate imaging at 20 minutes postinjection.
However, they are not usually seen on I-123 scans imaged
at 4 hours because the radiopharmaceutical has cleared.
Higher generalized background is seen on Tc-99m
pertechnetate compared to I-123 imaging.
Esophageal activity can be problematic with either
agent. It is frequently not in the midline, being displaced
by the trachea and cervical spine when the neck is
hyperextended in the imaging position. It is more often
Figure 5-15 Esophageal activity.Anterior (ANT) and right and left anterior oblique (RAO, LAO)
views obtained with Tc-99m pertechnetate. Note the esophageal activity below the thyroid to the
left of midline (arrows).
seen just to the left of midline and can usually be con-
firmed by having the patient swallow water to clear the
esophagus, followed by repeat imaging (Fig. 5-15).
Abnormal Thyroid Scintigraphy
A systematic interpretation of the thyroid scintigram
requires assessment of thyroid size and configuration and
the identification of focal abnormalities. These include
hot and cold nodules and extrathyroidal activity in the
neck or mediastinum. The thyroid scan allows for corre-
lation of palpable abnormalities with scintigraphic find-
ings. This is frequently critical in assigning significance
to a palpable abnormality. Radionuclide markers can aid
in confirming that the palpated nodule correlates with
the scintigraphic finding.
Gland size can be estimated using the thyroid scan but
this has limitations due to the scan’s two-dimensional
nature and the magnification and distortion caused by
pinhole collimation. However, image appearance and sur-
face radiomarkers provide some indication of size.
Enlargement seen on the scan is often accompanied by
a change from a relatively concave to a convex appear-
ance of the lobes.
The major clinical applications for thyroid scintigra-
phy are listed in Box 5-2. The specific scan findings seen
in various disease processes are discussed in this section.
Thyroid Nodule
Thyroid nodules are quite common. They occur more
often in women than men. The incidence of both benign
and malignant nodules increases with age. Determining
whether a nodule is benign or malignant is a common clin-
ical problem. A nodule presenting in a young person,
a male,or with recent nodule growth increases concern for
malignancy. The presence of multiple nodules decreases
the likelihood of malignancy. A nodule in a patient with
Graves’disease requires evaluation (Fig.5-16).

Figure 5-16 Cold thyroid nodule and Graves’ disease. Patient
presented with a palpable nodule in the inferior aspect of the left
lobe.TSH was suppressed and T4 elevated. I-123 thyroid scan
shows a non-functional nodule.The gland is full and the pyramidal
lobe prominent.The nodule must be evaluated prior to radioiodine
therapy.
Radiation to the head and neck or mediastinum is
associated with an increased incidence of thyroid cancer,
particularly in children. Several decades ago, external
radiation therapy was used to shrink asymptomatic
enlarged thymus glands and to treat enlarged tonsils, ade-
noids, and acne. This radiation therapy, usually in the
range of 10–50 rads, was associated with an increased
incidence of thyroid cancer. The radiation released at
Hiroshima, Nagasaki, and Chernobyl also resulted in an
increased incidence of thyroid cancer.
Radiation exposure up to 1500 rads increases the inci-
dence of thyroid nodules and cancer. The mean latency
period is approximately 5 years. For radiation greater than
1500 rads, the risk decreases, presumably due to tissue
destruction. High doses of radiation used in the therapy of
malignant tumors are more likely to cause hypothyroidism.
Ultrasonography
Nodules can be confirmed on sonography when sus-
pected on physical exam. Sonography can also detect
additional nodules to the palpated one and determine
whether a nodule is solid or cystic. Purely cystic lesions
are benign; however, cancer cannot be excluded if the
cyst has a soft tissue component or cystic degeneration.
Fine Needle Aspiration (FNA)
Aspiration biopsy is often performed today without prior
scintigraphy because it is a more direct means of estab-
lishing the histology of solitary nodules. This procedure
is often performed in the endocrinologist’s office.
Endocrine System 89
Figure 5-17 Toxic (hot) thyroid nodule. Patient presented with
suppressed TSH, elevated T4 and a palpable nodule in the right
lobe of the thyroid. Uptake in the nodule is intense and the
remainder of the gland is suppressed.
Sonographic-guided biopsy is used for nonpalpable nod-
ules. Although aspiration biopsy is subject to sampling
error, the overall accuracy is high. Benign follicular neo-
plasms cannot always be distinguished histopathologi-
cally from follicular cancer.
Thyroid Scintigraphy
The thyroid scan is more sensitive than physical examina-
tion for detecting nodules. Although ultrasonography is
more sensitive than scintigraphy for detection of small
nodules, the natural history and clinical significance of
small nodules seen only on sonography is uncertain. The
data used to assign risk to hot or cold nodules is based on
scintigraphy correlation with pathology.
The thyroid scan does not diagnose nodules per se.
A hot or cold region on the scan may be due to various
other pathologies discussed later (e.g., thyroiditis and
scarring). A nodule is diagnosed by physical examina-
tion of the thyroid or detected by an anatomical imaging
modality (e.g., ultrasonography, CT, or MRI).
Thyroid scintigraphy can determine the functional sta-
tus of a nodule detected by physical examination or
anatomical imaging. Thyroid nodules are classified scinti-
graphically as cold (hypofunctioning compared to adjacent
normal tissue) (Figs. 5-11 and 5-16), hot (hyperfunctioning
with suppression of the extranodular gland) (Fig. 5-17),
warm (increased uptake compared to adjacent tissue but
without suppression of the extranodular tissue (Fig. 5-18),
or indeterminate (palpable but not visualized on scintigra-
phy). The scan can also show the presence of multiple nod-
ules (multinodular goiter) (Fig. 5-19). This interpretation
90 NUCLEAR MEDICINE: THE REQUISITES

system can provide a relative risk assessment for malig-

nancy (Box 5-10).
Cold Nodule
Greater than 85–90% of thyroid nodules are cold (hypo-
functional) on thyroid scintigraphy, that is, they have
decreased uptake compared to adjacent thyroid tissue.
Many have benign etiologies, such as simple cysts, colloid
nodules, thyroiditis, hemorrhage, necrosis, and infiltrative
disorders such as amyloid or hemochromotosis (Box
5-11). However, a significant subgroup of patients with
cold nodules has malignancy as the etiology.

Box 5-10 Likelihood of Thyroid Cancer

in Nodule Based on Thyroid
Scintigraphy

NODULE LIKELIHOOD OF THYROID CANCER

Cold 15–20%
Figure 5-18 Warm thyroid nodule in euthyroid patient. Patient Indeterminate 15–20%
presented with a palpable 1.5 cm nodule. Increased uptake is Multinodular 5%
seen in the inferior aspect of the right lobe of the thyroid. The Hot <1%
extranodular gland is not suppressed. The patient had normal
thyroid function tests. Thus, the nodule may be autonomous but it
is not a toxic nodule.

Box 5-11 Differential Diagnosis for

Thyroid Nodules

Cold nodules (non-functioning)

Benign
Colloid nodule
Simple cyst
Hemorrhagic cyst
Adenoma
Thyroiditis
Abscess
Parathyroid cyst or adenoma
Malignant
Papillary
Follicular
Anaplastic
Medullary
Hürthle cell
Lymphoma
Metastatic carcinoma
Lung
Breast
Melanoma
Gastrointestinal
Figure 5-19 Multinodular goiter. The patient has multiple Renal
thyroid nodules and thyromegaly by examination. There are Functioning nodules (warm or hot)
increased and decreased regions of uptake in full (convex borders) Adenomas
gland. The patient is euthyroid. The extrathyroidal tissue is not Hyperfunctioning adenomas
definitely suppressed.
 Endocrine System 91

The incidence of thyroid carcinoma in a single cold

nodule is reported to be as high as 40% in surgical series
but as low as 5% in general medical series. Overall, the
incidence of cancer in a cold thyroid nodule is generally
considered to be approximately 15–20%.With multinodu-
lar goiters, the incidence of malignancy in cold nodules is
lower, less than 5%. Enlarging nodules or “dominant” nod-
ules (i.e.,those that are distinctly larger than the other nod-
ules in a multinodular goiter) require further evaluation
because of relatively increased risk.
Hot and Warm Nodules
Radioiodine uptake within a nodule denotes function.
A functioning nodule is very unlikely to be malignant.
Less than 1% of hot nodules harbor malignancy. The
term hot nodule should be reserved for those that not
only have high uptake in the nodule scintigraphically, but
also have suppression of extranodular tissue (see Fig. 5-
17). If extranodular tissue is not suppressed, it should be
referred to as a warm nodule.
Hot nodules are caused by toxic adenomatous nodules.
Warm nodules may be caused by autonomous hyperfunc-
tioning adenomas. However, they are not toxic, that is, they
are not producing enough thyroid hormone to cause thyro-
toxicosis and thus TSH is not suppressed. A warm nodule
may also be due to nonautonomous hyperplastic tissue or
even normal functioning tissue surrounded by poorly func-
tioning thyroid. Differentiation can be made by administra-
tion of thyroid hormone (thyroid suppression test).
Autonomous nodules cannot be suppressed (see T-3
Suppression Test). However, the suppression test is rarely
needed in current practice.
Large hot nodules greater than 2.5–3.0 cm usually pro-
duce overt hyperthyroidism. Some patients with smaller
nodules have subclinical hyperthyroidism, which can be
confirmed by a suppressed serum TSH but normal T4. In
the past, a small autonomous nodule might be followed
clinically because some stabilize, whereas others regress
or undergo involution (Fig. 5-20). Increasingly, nodules
are treated at an early stage because of the low incidence Figure 5-20 Spontaneous resolution of a hot nodule.
A, Thyroid scan reveals a large, hot nodule in the left lobe of the
of regression and increased awareness of adverse conse- thyroid.The center of the nodule appears to have less intense
quences associated with subclinical hyperthyroidism tracer activity than the periphery, suggesting central degeneration.
(e.g., bone mineral loss). B, Follow-up scan 1 year later reveals complete involution of the
Radioiodine I-131 is the usual therapeutic method of hot nodule, with residual distortion of the gland.
choice for toxic nodules. Radiation is delivered selec-
tively to the hyperfunctioning tissue while sparing sup-
pressed extranodular tissues. The suppression of normal by thyroid scan as hot or cold compared to surrounding
tissue results in a low incidence of posttherapy hypothy- normal thyroid, it is referred to as an indeterminate nod-
roidism. After successful treatment of the nodule, the ule. A cold nodule arising from the posterior aspect of the
suppressed tissue regains function.Surgery,usually lobec- gland may have normal glandular activity superimposed
tomy, may be indicated if there are local symptoms or over the nodule, making it appear to have normal uptake.
cosmetic concerns. For management purposes, an indeterminate nodule has
Indeterminate Nodule the same significance as a cold nodule. The possibility of
When a palpable or sonographically detected nodule an indeterminate nodule highlights the need for close cor-
greater than a centimeter in size cannot be differentiated relation between physical and scintigraphic findings.
92 NUCLEAR MEDICINE: THE REQUISITES

Discordant Nodule Still, a minority of discordant nodules are malignant

Discordance in appearance between radioiodine and
Tc-99m pertechnetate scans is seen in a small minority
of patients. A nodule may appear hot on pertechnetate
imaging but cold on radioiodine imaging because Tc-99m
pertechnetate is trapped but not organified (Fig. 5-21).
This discordance occurs in approximately 5% of hot nod-
ules seen on Tc-99m pertechnetate scans. Because some
thyroid cancers maintain trapping but not the organifica-
tion function, a single hot nodule identified on Tc-99m
pertechnetate imaging should not be considered a func-
tioning nodule until confirmed by radioiodine studies.
(20%). The discordant nodule is a drawback to the use of
Tc-99m pertechnetate for routine thyroid scintigraphy.
Goiter
The term goiter refers to thyroid gland enlargement, but
it is often qualified to indicate the cause of the enlarge-
ment (e.g., toxic nodular goiter, colloid goiter, or diffuse
toxic goiter [Graves’ disease]) (Box 5-12).
Graves’ Disease vs. Multinodular Toxic Goiter
In a patient with newly diagnosed thyrotoxicosis, the
physical exam can usually differentiate the diffuse goiter

Figure 5-21 Discordant nodule. A, Tc-99m pertechnetate scan reveals a functioning nodule in
the left upper pole. B, In the corresponding iodine-131 scintigram the nodule is cold. Thus the
nodule can trap but not organify iodine. This requires further work-up.

Box 5-12 Conditions Associated with Goiter

Graves’ disease: Autoimmune disease associated with hyperthyroidism and exophthalmos. Patients have diffuse hyperplasia
of the thyroid gland and thyroid-stimulating immunoglobulins (TSIs).
Plummer’s disease: Hyperthyroidism associated with toxic nodular goiter (one or more nodules).
Hashimoto’s disease (Hashimoto’s thyroiditis): Form of thyroiditis and autoimmune disorder often leading to
hypothyroidism. Patients may rarely experience transient hyperthyroidism (“Hashitoxicosis”).
de Quervain’s disease: Subacute thyroiditis, with granulomatous infiltration and destruction of thyroid cells.Transient
hyperthyroidism early in course.
Riedel’s struma (Riedel’s thyroiditis): Chronic fibrous replacement of the thyroid gland.
Jod-Basedow phenomenon: Induction of thyrotoxicosis in a euthyroid individual after exposure to large amounts of iodine.
Occurs in areas of endemic iodine-deficient goiter or after use of iodine contrast agents.
Wolff-Chaikoff effect: Paradoxical blocking of iodine incorporation into thyroid hormone resulting from large amounts of
iodine.
Marine-Lenhart syndrome: Graves’ disease with incidentally functioning nodules that are responsive to thyroid-stimulating
hormone but are not responsive to TSIs.

of Graves’ disease (Fig. 5-22) from a multinodular toxic
gland (see Fig. 5-19). The thyroid scintigram can help
make the distinction. Toxic nodular goiter has the char-
acteristic scintigraphic pattern of increased uptake that
corresponds to palpable nodules and suppression of
extranodular thyroid tissue. This contrasts with the dif-
fuse homogenous increased uptake of Graves’ disease.
The pyramidal lobe, a paramedian structure arising supe-
riorly from the isthmus (right or left lobe), is also usually
well visualized (see Fig. 5-22) with Graves’ disease.
Colloid Nodular Goiter
Before the addition of iodine supplements to salt and
food, goiter was endemic in the northern United States
around the Great Lakes, and still occurs in some parts of
the world. These endemic goiters were typically com-
posed of colloid nodules (colloid nodular goiters) and
most were benign.
The pathogenesis of nodule formation in these
patients is iodine deficiency-induced hyperplasia fol-
lowed by the formation of functioning nodules that
undergo hemorrhage and necrosis replaced by lakes of
colloid. Repetition of this process over time leads to
glandular enlargement, with nonfunctioning colloid nod-
ules as the dominant histopathological feature. The typi-
cal scintigraphic appearance of benign multinodular
colloid goiters is inhomogeneous uptake of tracer with
cold areas of various sizes (Fig. 5-23).
Substernal Goiter
Most substernal goiters are extensions of the thyroid into
the mediastinum. As they enlarge, they may cause symp-
Figure 5-22 Graves’ disease. Large goiter with high uptake.
The %RAIU was 65%. Note the pyramidal lobe.
Endocrine System 93
toms of dyspnea, dysphagia, or stridor. Many are asymp-
tomatic and incidentally detected as an anterior upper
mediastinal mass on CT. A radioiodine scan can confirm
the thyroid origin of this mass (see Fig. 5-10).
Scintigraphy Tc-99m pertechnetate is not suited
for this purpose because of its high mediastinal blood
pool activity. The ability to perform delayed imaging
after tissue and blood pool clearance of background
activity is the advantage of I-131 for evaluation of a
substernal goiter. Iodine uptake in substernal goiters is
often poor and the highest target-to-background ratio
possible is desirable.Imaging at 24–48 hours is sometimes
necessary.
The rationale for the use of I-131 rather than I-123 for
evaluating a substernal goiter has been that I-123 would
be attenuated by the sternum. However, attenuation with
I-123 is only in the range of 10–20%. In most cases, I-123
can provide similar information with better image quality
and lower radiation to the patient.
The usual cervical location of the thyroid gland should
always be imaged when searching for a substernal goiter
because most demonstrate continuity with the cervical
portion of the gland, although some patients have only
a fibrous band connecting the substernal and cervical
thyroid tissues.
Ectopic Thyroid Tissue
The thyroglossal duct runs from the foramen caecum at
the base of the tongue to the thyroid. If it fails to migrate
Figure 5-23 Colloid nodular goiter. Goiter with
inhomogeneous tracer distribution and focal cold areas
corresponding to nodules.
94 NUCLEAR MEDICINE: THE REQUISITES
from its anlage, lingual or upper cervical thyroid tissue
can present in the neonate or child as a midline mass
with or without obstructive symptoms and often accom-
panied by hypothyroidism. Ectopic thyroid tissue may
also be mediastinal (substernal goiter) or even pelvic/ovar-
ian (struma ovarii).
Scintigraphy
The typical appearance of a lingual thyroid is a focal or
nodular accumulation at the base of the tongue and
absence of tracer uptake in the expected cervical loca-
tion (Fig.5-24).However,lingual thyroids usually function
poorly. Lateral thyroid rests are also often hypofunc-
tional. However, rests can function, hyperfunction, or be
involved with adenocarcinoma of the thyroid. Functioning
ectopic thyroid tissue should be considered metastatic
until proven otherwise.
Subacute Thyroiditis
This entity is discussed in the section Thyrotoxicosis.
With hyperthyroidism, the scan shows only suppression
(see Fig. 5-13). During recovery phases, the appearance of
the thyroid is variable and depends upon the severity and
distribution of the disease (Fig. 5-25). The scintigram
may show inhomogeneity of uptake, regional areas of
hypofunction, or even focal hypofunction.
Chronic Thyroiditis (Hashimoto’s Thyroiditis)
Scintigraphic findings are highly variable.Diffuse enlarge-
ment is usual, although the scan may be normal early in
the process. Uptake may be inhomogeneous throughout
the gland or there may be focal cold areas without a palpa-
Figure 5-24 Lingual thyroid. Hypothyroid infant with neck mass. Scan (anterior view) shows
prominent uptake within the neck mass and no thyroid in region of thyroid bed.
ble nodule. The pyramidal lobe is often seen in
Hashimoto’s disease.
Acute Thyroiditis
Suppurative bacterial infection is the usual cause for this
rare condition. The thyroid is typically enlarged and ten-
der. Focal abscesses will appear as cold regions scinti-
graphically. Reidel’s struma is an uncommon form of
thyroiditis where all or part of the gland is replaced by
fibrous tissue. No uptake is seen in the region of fibrous
tissue.
Thyroid Cancer
Whole body thyroid cancer scintigraphy has long been
used for well-differentiated papillary and follicular
thyroid cancer. It is often performed post-thyroidectomy
prior to radioiodine I-131 therapy and for evaluating
response therapy (Fig. 5-26). The most common sites of
metastasis are locally in the lymph nodes of the neck,
lung, mediastinum, and bones (Fig. 5-27). Medullary car-
cinomas and anaplastic carcinomas do not concentrate
radioiodine and are not detected with conventional thy-
roid scintigraphy.
Whole body thyroid cancer scanning requires patient
preparation. The traditional approach is to withdraw
hormone replacement therapy for 4–6 weeks so that
patients may achieve a maximal endogenous thyroid-
stimulating hormone (TSH) response (>30 U/ml). To
minimize symptoms of hypothyroidism,patients are some-
times switched to short-acting triiodothyronine (T3),
which is discontinued 2 weeks prior to the scan.
 Endocrine System 95

Figure 5-25 Resolving subacute thyroiditis. A, Tc-99m pertechnetate scan in a patient with
resolving subacute thyroiditis affecting the left lobe.The patient is now moderately hypothyroid.
There is lack of tracer uptake in the left lobe and decreased accumulation in the right lobe.
B, Corresponding gallium-67 scintigram reveals marked focal accumulation in the area of the left
lobe, indicating inflammatory nature of the process.

Imaging with I-131 is typically performed 48 hours
after I-131 diagnostic dose administration (Box 5-13).
More lesions are demonstrated in this time than at 24
hours due to background clearance and the higher tar-
get-to-background ratio. Serum thyroglobulin levels are
also measured during maximum TSH stimulation (a sensi-
tive tumor marker). For I-123, whole body imaging is
acquired at 24 hours (Box 5-14).
Postthyroidectomy, it is not uncommon to have high
intensity uptake in the thyroid bed (star artifact), which
may preclude good visualization of the neck or medi-
astinum. The artifact is caused by septal penetration of
high-energy photons through the collimator (see Fig.
5-12). A pinhole collimator that has no septa can better
resolve the high intensity uptake in the neck. In the post-
operative state,uptake in the neck may be due to residual
normal thyroid or to thyroid cancer. The scan cannot
make the distinction. Activity outside the thyroid bed is
very likely metastatic.
Other Thyroid Imaging
Radiopharmaceuticals
Tl-201 and Tc-99m Sestamibi
Both radiotracers are nonspecific tumor imaging agents.
Uptake of these radiopharmaceuticals occurs in benign
as well as malignant conditions. Thus, they have not
found widespread application in the initial diagnosis of
thyroid cancer. Some advocate their use for thyroid can-
cer follow-up imaging, where differentiating uptake in
normal tissue and benign lesions from tumor is not an
issue. One advantage is that patients need not discon-
tinue thyroid hormone replacement therapy before
imaging. This approach has mainly found acceptance for
locating metastases in patients with increased serum thy-
roglobulin levels and negative radioiodine whole body
scintigrams. F-18 fluorodeoxyglucose (FDG) positron
emission tomography (PET) imaging is rapidly taking
over that indication.
F-18 Fluorodeoxyglucose (FDG)
The sensitivity of F-18 FDG PET for detection of well-dif-
ferentiated thyroid cancer metastases is not high (~70%)
and thus not generally used for routine thyroid cancer
imaging. Its major indication has been in patients who
have had negative I-131 whole body scans but elevated
serum thyroglobulin levels. In these patients, the tumor
has dedifferentiated into a higher grade tumor. Many of
these patients have focal uptake of F-18 FDG at the site of
metastasis (Fig. 5-28). A higher-grade tumor increases the
likelihood of FDG uptake.Localization of the tumor allows
for surgical resection or evaluation of response to therapy.
On whole body FDG PET performed for staging or
surveillance of other oncologic tumors, focal F-18 FDG
uptake is occasionally seen in the thyroid. This find-
ing usually indicates thyroid pathology. Up to one-half
of these patients have incidental primary thyroid can-
cers diagnosed. Other causes of focal uptake include
96 NUCLEAR MEDICINE: THE REQUISITES
metastatic cancer, benign follicular adenoma, and thy-
roiditis. Diffuse gland uptake can be seen with thy-
roiditis and Graves’ disease.
I-131/I-123 Metaiodobenzylguanidine (MIBG)
Some medullary carcinomas of the thyroid demonstrate
I-123 or I-131 MIBG uptake. However, the sensitivity is
low (~30%). MIBG localizes in neurosecretory storage
vesicles of chromaffin cells. Soft tissue metastases are bet-
ter visualized than skeletal metastases. The low sensitiv-
ity precludes a routine role for MIBG in the workup of
medullary thyroid cancer.
Figure 5-26 Whole body I-123 thyroid
cancer scans pre- and post-I-131 ablation
therapy. A, The postthyroidectomy, pre-I-
131 ablation therapy scan shows uptake
limited to three focal areas in the thyroid
bed. There are no local or distant
metastases. Normal stomach and urinary
clearance. B, Seven days post-I-131
therapy, there has been no significant
change from the pretherapy scan,
except for liver uptake due to metabolism
of I-131 labeled thyroid hormone.
Indium-111 Somatostatin Receptor Scintigraphy
Medullary carcinoma of the thyroid is a neuroectodermal
tumor. However, unlike many neuroendocrine tumors,
sensitivity for detection of medullary carcinoma of the
thyroid with In-111 somatostatin receptor scintigraphy
(OctreoScan) is low (<50%).
Other Thyroid Function Studies
Most of these classical pharmacologic interventions
are not commonly used today and are of historical
interest, but are pertinent to an understanding of
 Endocrine System 97

Figure 5-27 Lung metastases seen post-I-131 therapy. Follicular thyroid carcinoma treated with
total thyroidectomy and I-131 ablation three years prior. Serum thyroglobulin level is now elevated.
A, I-123 whole body pre-therapy scan shows no uptake in the neck or elsewhere to suggest
metastases.
Continued

thyroid physiology and interpretation of present-day
studies.
T3 Suppression Test
The test’s clinical utility was for diagnosing patients with
borderline Graves’ disease and autonomous functioning
glands. In the T3 suppression test, a baseline 24-hour
uptake is obtained. The patient then receives 25 mcg of
T3 four times a day for 8 days. The 24-hour uptake is
repeated beginning on day 7. A normal response to thy-
roid suppression is a fall in the percentage of uptake to
less than 50% of the baseline value and less than 10%
overall. An autonomously functioning gland will not
suppress. Very sensitive tests for TSH levels are now used
and can accurately detect early hyperthyroidism.
TSH Stimulation Test
The test has been used to distinguish primary from sec-
ondary (pituitary) hypothyroidism. Failure to respond to
exogenous TSH is indicative of primary hypothyroidism.
98 NUCLEAR MEDICINE: THE REQUISITES

Figure 5-27, cont’d B, After 150 mCi of I-131, the scan shows uptake diffusely throughout both
lung fields consistent with miliary lung metastases.

Patients with secondary hypothyroidism have increased
radioiodine uptake after TSH stimulation. The stimula-
tion test is performed by first determining a baseline 24-
hour radioiodine percent uptake. The patient then
receives TSH intramuscularly. The %RAIU is repeated
beginning the next day. In healthy subjects and patients
with hypopituitarism the uptake should double, whereas
those with primary hypothyroidism show no response.
Perchlorate Discharge Test
The perchlorate discharge test demonstrates dissocia-
tion of the trapping and organification functions in the
 Endocrine System 99

Box 5-13 Iodine-131 Whole Body

Imaging for Thyroid Cancer:
Protocol Summary

PATIENT PREPARATION
Discontinue thyroid hormone for a sufficient period (T4
for 6 weeks,T3 for 2 weeks) to ensure maximum
endogenous thyroid-stimulating hormone response
(>30 μU/mL).

RADIOPHARMACEUTICAL
Withdrawal: 2 mCi (74 MBq), orally
Thyrogen: 4 mCi (148 MBq)

IMAGING TIME
At 48 hours.

PROCEDURE
Use a wide field-of-view gamma camera with computer
acquisition.
Use a high-energy parallel-hole collimator and a 20% Figure 5-28 F-18 FDG PET scan for thyroid cancer. Patient had
window centered at 364 keV. negative I-123 whole body scan but elevated serum thyroglobulin.
The FDG PET scan shows focal uptake in the left neck as well as
Whole body scan and a 20-min spot view to include
abnormal uptake in the paratracheal region, consistent with
head, neck, and mediastinum. dedifferentiated tumor metastases.
Calculate a percent radioactive iodine uptake.

thyroid. This occurs with congenital enzyme deficien-

cies, in chronic thyroiditis, and during therapy with
propylthiouracil. The patient receives a tracer dose of
radioiodine. The percent uptake is measured at 1–2
Box 5-14 Iodine-123 Whole Body hours. One gram of potassium perchlorate is then given
Imaging for Thyroid Cancer: orally and the percent uptake is measured hourly.
Protocol Summary A washout greater than 10% suggests an organification
defect.
PATIENT PREPARATION
Discontinue thyroid hormone for a sufficient period (T4 Radioiodine Therapy
for 6 weeks,T3 for 2 weeks) to ensure maximum Graves’ Disease
endogenous thyroid-stimulating hormone response
Pharmacologic and Surgical Therapy
(>30 μU/mL).
Patients with newly diagnosed Graves’ disease are often
RADIOPHARMACEUTICAL treated with beta-blockers, which provide prompt
Withdrawal: 1.5 mCi (56 MBq), orally symptomatic relief. More specific therapy with
Thyrogen: 2 mCi (74 MBq) thiourea antithyroid drugs that block organification,
such as propylthiouracil (PTU) and methimazole
IMAGING TIME (Tapazole), reduce thyroid hormone production.
At 24 hours. These drugs are used to “cool” the patient down, render
them euthyroid, and provide time to consider further
PROCEDURE
therapeutic options. Patients may take these drugs for
Use a wide field-of-view gamma camera with computer 6–12 months. These drugs have a rather high inci-
acquisition.
dence of adverse effects (50%), the most serious being
Use a high-energy parallel-hole collimator and a 20%
window centered at 364 keV.
liver dysfunction and agranulocytosis. The latter side
Whole body scan and a 20-min spot view to include effects are not common but quite serious, thus the
head, neck, and mediastinum. drugs are rarely administered for longer than a year.
Calculate a percent radioactive iodine uptake. Thyroidectomy is a rather uncommon form of therapy
for Graves’ disease.
100 NUCLEAR MEDICINE: THE REQUISITES
Radioactive Iodine
Most patients with Graves’ disease are treated with
radioactive iodine I-131, some early after diagnosis and
others 6–12 months later (Box 5-15). There is over 50
years of experience with use of therapeutic I-131.
Endocrinologists have become comfortable with treating
patients,even children,with I-131 because of its high effi-
cacy and low incidence of adverse affects.
Most patients with Graves’ disease are effectively
treated with one therapeutic dose of I-131. Approxi-
mately 10% require a second treatment. Symptomatic
improvement is usually noted by 3 weeks after therapy.
However, the full therapeutic effect takes 3–6 months
because stored hormone must first be released and used.
The exophthalmos of Graves’ disease is not controlled by
thiourea drugs or radioactive iodine. Some evidence sug-
gests exacerbation of exophthalmos with I-131 therapy,
so steroids are often administered.
In women, pregnancy must be ruled out before I-131
therapy. Because the fetal thyroid begins concentrating
iodine at 10–12 weeks of gestation, cretinism may occur
after therapeutic doses of I-131 given during pregnancy.
Women should be counseled to avoid pregnancy for 3–6
months after therapy in case retreatment is indicated.
Most patients eventually become hypothyroid and
need replacement hormone therapy. Hypothyroidism
can occur within several months or decades after ther-
apy. A greater administered dose causes earlier onset of
hypothyroidism, and vice versa.
Box 5-15 Indications for Iodine-131
Therapy
INDICATED
Graves’ disease (diffuse toxic goiter)
Plummer’s disease (toxic nodular goiter)
Functioning thyroid cancer (metastatic)
NOT INDICATED
Thyrotoxicosis factitia
Subacute thyroiditis
“Silent”thyroiditis (atypical, subacute, lymphocytic,
transient, postpartum)
Struma ovarii
Thyroid hormone resistance (biochemical/clinical
manifestations)
Secondary hyperthyroidism (pituitary tumor, ectopic
thyroid-stimulating hormone, trophoblastic tumors
[human chorionic gonadotropin])
Thyrotoxicosis associated with Hashimoto’s disease
(“Hashitoxicosis”)
Jod-Basedow phenomenon (iodine-induced
hyperthyroidism)
Various approaches have been used for selecting an I-131
dose for therapy of Graves’ disease. One approach is to pre-
scribe a standard I-131 dose in the range of 8–15 mCi.
Overall,this usually works well.However,factors such as the
size of the gland and the %RAIU may result in very different
radiation doses to the thyroid in different patients.Generally,
large glands require a relatively higher therapeutic dose and
patients with a high %RAIU require a lower dose.
A commonly used formula takes both of these factors,
gland size and %RAIU, into consideration (Box 5-16):
gram size of thyroid gland
× 100–180 μCi/gm
I-131 administered dose =
24-hour %RAIU
This approach calculates an individual therapy dose for
each patient with Graves’ disease. An estimation of the
gram weight of the gland is required. A normal gland
weighs 15–20 gm. Patients with Graves’ disease often
have glands in the range of 40–80 gm and larger. There is
considerable interphysician variability in gland size esti-
mation.
The other variable in this calculation is the μCi/gm
dose. Often in the past, referring physicians requested
low I-131 doses to minimize the radiation to the patient
(e.g., 60–80 μCi/gm tissue). Referring physicians are
increasingly comfortable with the safety of higher doses
(120–180 μCi/gm tissue) and prefer the increased likeli-
hood of success with a single therapeutic dose. Early
onset hypothyroidism is also often preferred by some
because it is inevitable in any event and prompt replace-
ment therapy can be instituted.
In patients demonstrating rapid radioiodine turnover in
the gland (i.e., high 4-hour but normal or near-normal
24-hour %RAIU), a higher I-131 dose than would be
Box 5-16 Calculation of Iodine-131
Therapeutic Dose for Graves’
Hyperthyroidism
INPUT DATA
Gland weight: 60 g
24-hour uptake: 80%
Desired dose to be retained in thyroid (selected to
deliver 8000 to 10,000 rads to thyroid): 100 μCi/g
CALCULATIONS
60 g × 100 μCi/g
Required dose (μCi) = = 7500
0.80
Dose (mCi) = 7500 = 7.5 mCi
1000

calculated using the 24-hour %RAIU should be considered.
This indicates a shorter residence time within the thyroid.
Some patients report local neck pain, tenderness, and
swelling after I-131 therapy due to radiation thyroiditis.
Thyroiditis results in release of thyroid hormone and poses
the risk of thyroid storm,although this is rare.Patients with
florid disease and those treated with higher amounts of
radioactivity are at greater risk. In older patients who have
preexisting heart disease and in patients otherwise at risk,
medical therapy with thiourea drugs can be carried out for
several months to deplete thyroid hormone before radioio-
dine therapy. Beta blockers are often used both before and
after therapy to minimize this risk.
Other secondary effects of radiation exposure (e.g.,sec-
ondary cancers, infertility, abnormal offspring) have been
a concern. Evidence collected over a half-century of I-131
therapy has shown no statistically significant difference in
the frequency of these events between patients receiving
I-131 therapy and patients treated by surgery for Graves’
disease.I-131 therapy does not reduce fertility and congen-
ital defects are not increased in the children of treated indi-
viduals. Patients treated with I-131 have a somewhat
higher incidence of leukemia than the general population.
Toxic Nodular Goiter
Patients with hyperthyroidism caused by a toxic multi-
nodular goiter are generally more resistant to therapy
with radioiodine than patients with diffuse toxic goiter
(Graves’ disease). The reason is uncertain. Radioiodine
turnover may be higher in these nodules, leading to
a lower retained dose. Thus the dose of radioiodine
should be increased by at least 50% over what would be
given for Graves’ disease. An empirical dose, often in the
range of 20–25 mCi, is often given. Because extranodular
tissue in the thyroid is suppressed, it is spared from radia-
tion and usually resumes normal function after success-
ful therapy. Patients with a single toxic nodule are often
treated similarly. A more sophisticated approach would
be to use the formula described for Graves’ disease (see
Box 5-16). In place of the estimated gland size for Graves’
disease, replace it with a calculation of the volume of the
nodule (V = 4/3 Πr3).
Thyroid Cancer
Radioactive iodine I-131 has also been used extensively
in the treatment of differentiated thyroid cancer. It is not
useful for treating anaplastic and medullary cancers.
Postsurgical ablation of normal thyroid remnants reduces
local recurrences and allows the patient to be followed
with serum thyroglobulins and radioiodine whole body
thyroid cancer scans. Patients with residual or recurrent
differentiated thyroid cancer have improved survival
with I-131 treatment.
Patients are prepared for therapy by discontinuing
thyroid hormone therapy. A diagnostic scan is usually
Endocrine System 101
done to establish the presence of residual thyroid tissue
or metastatic disease. Serum thyroglobulin levels are also
measured. In the past, patients had to remain hospitalized
and isolated until retained whole body activity was less
than 30 mCi. The Nuclear Regulatory Commission has
published new rules (10CFR 20 and 35) for release of
patients based on the likely exposure to others. The
release criterion state that no person should receive more
than 5 millisieverts (0.5 rem) from exposure to a released
I-131 treated patient. In general, this allows for consider-
ably higher release activity. At some centers, the majority
of patients are now treated on an outpatient basis.
Radiation safety instructions are discussed with the
patient and family. Patient specific information regarding
limiting close contact and preventing exposure to others
is provided.
A whole body scan 7 days after the therapeutic dose is
generally routine. Metastatic disease not seen on the
pretherapy diagnostic scan may be detected due to the high
therapy administered dose. After therapy, the patient is
placed back on thyroid hormone replacement and sup-
pressive therapy. Retreatment is usually not considered
for at least 6 and usually 12 months to avoid bone mar-
row damage.
Metastatic disease most commonly occurs locally in
the neck. Distant metastases have a predilection for the
lung and skeleton. Brain and liver metastases are less
common. Therapeutic doses vary somewhat by institu-
tion; however, regional neck metastases typically receive
75–100 mCi and lung and bone metastases receive
150–200 mCi. Follow-up imaging is usually carried out at
yearly intervals until all detected metastases are elimi-
nated with repeat therapy. Concern about bone marrow
suppression and leukemia increases as the total dose
approaches 500 mCi. Doses greater than 1000 mCi are
rarely administered.
PARATHYROID SCINTIGRAPHY
Parathyroid scintigraphy has changed over the years from
a procedure with a variable reported accuracy that was
considered unnecessary by many surgeons to a much
improved imaging methodology with good accuracy,
which has become accepted by endocrinologists and
neck surgeons as part of the routine preoperative
workup of hyperparathyroidism. The importance of pre-
operative parathyroid imaging has paralleled the growth
of minimally invasive surgery and the use of intraopera-
tive gamma probe methodology.
Anatomy and Embryology
There are usually four parathyroid glands, two upper and
two lower, measuring approximately 6 mm × 3 mm
102 NUCLEAR MEDICINE: THE REQUISITES
and weighing 35–40 gm. A fifth supernumerary gland
occurs in 10% of individuals. Rarely, there may be only
two glands or as many as eight.
The inferior parathyroid glands arise from the third
branchial pouch and migrate caudally with the thymus.
Their location is somewhat variable with 60% located just
posterior to the lower thyroid poles, 25% in the cervical
portion of the thymus gland,and a minority within the thy-
roid gland or at various positions from the angle of the jaw
to the arch of the aorta (Fig. 5-29).The superior glands
arise from the fourth branchial pouch and migrate with
the thyroid. They are usually just posterior to the upper
thyroid lobes. Only 1% are ectopic, located between the
thyroid and esophagus, within the carotid sheath, behind
the innominate vein,or in the posterior mediastinum.
Figure 5-29 Normal and aberrant location of parathyroid
glands.The superior pair of glands (striped circles) often lie within
the fascial covering of the posterior aspect of the thyroid gland
outside the capsule, although rarely intrathyroidal. Most are
adjacent to the thyroid or cricothyroid cartilage, rarely
retropharyngeal or retroesophageal. Inferior glands (black circles)
are more variable. Many are located inferior, lateral or posterior to
the lower pole of the thyroid gland.They are commonly found in
the thyrothymic ligament or even in the cervical thymus.A small
percent migrate to the superior mediastinum. Rare ectopic glands
are found superiorly. Arrow indicates retraction of the thyroid.
There are two types of cells in the parathyroid. The pre-
dominant functioning cell is the chief cell which produces
parathormone (PTH). It has little cytoplasm or mitochon-
dria. The oxyphil cells have a higher proportion of mito-
chondria and are increased in number in diseased glands.
Pathophysiology
Hyperparathyroidism
Increased synthesis and release of parathyroid hormone
(PTH) from the chief cells of a single gland or multiple
glands characterizes this disease. PTH is an 84-amino acid
polypeptide synthesized, stored, and secreted by the
parathyroid glands. The hormone is responsible for cal-
cium and phosphorus homeostasis by its action on bone,
small intestine, and the kidneys. Sporadic parathyroid
adenomas are caused by somatic mutations with subse-
quent clonal expansion of the mutated cells. Primary
hyperplasia is a polyclonal proliferation. In patients with
familial multiendocrine neoplasia (MEN syndrome),
hyperparathyroidism is secondary to a multiglandular
hyperplasia.
Primary hyperparathyroidism is caused by abnormal
parathyroid gland or glands that secrete excessive PTH.
Greater than 85% of patients with primary hyperthy-
roidism have adenomas as the cause (Box 5-17). Most of
the others have multigland hyperplasia. Less than 1% of
patients with hyperparathyroidism have carcinoma of
the parathyroid gland. They tend to have marked eleva-
tions in serum calcium associated with palpable neck
mass, bone pain, fractures, and renal colic.
Secondary hyperparathyroidism occurs in patients
with renal insufficiency. It is an appropriate compensa-
tory mechanism to increase the serum calcium level,
which is low in patients with renal failure. Despite the
elevated serum PTH, the serum calcium level remains
below normal levels.
Tertiary hyperparathyroidism occurs in patients
with renal failure when one or more of the glands
become autonomous and produces hypercalcemia.
Clinical Presentation
In the past, patients often presented with nephrolithiasis
and osteitis fibrosa cystica associated with osteoporosis,
pathologic fractures, and brown tumors. Many had gas-
trointestinal and neuropsychiatric symptoms. Today,
Box 5-17 Etiology of Hyperparathyroidism
Adenoma 85%
Hyperplasia 10%
Ectopic <5%
Carcinoma <1%

most patients are asymptomatic and diagnosed during
routine blood test screening. The diagnosis is suspected
clinically because of an elevated serum calcium level and
a reduced serum inorganic phosphate. There are numer-
ous other causes of hypercalcemia. The most common is
malignancy. Others include vitamin D intoxication, sar-
coidosis,and thiazide diuretics.
Diagnosis
The diagnosis of primary hyperparathyroidism is made
clinically. An elevated parathormone level in a patient
with hypercalcemia is diagnostic of primary hyperparathy-
roidism. Most other causes of hypercalcemia, except
parathyroid carcinoma, have reduced parathormone lev-
els.Imaging is performed for localization,not diagnosis.
Treatment
Bone mineral density loss caused by hyperparathy-
roidism is of increasing concern and early surgical
resection is an increasingly common practice. The stan-
dard operation has been bilateral neck exploration
with removal of the adenoma.Hyperplasia usually requires
removal of 3.5 glands. Increasingly, minimally invasive sur-
gery is being performed, which permits a shortened oper-
ation time and fewer complications. The latter approach
requires preoperative imaging for localization. Some sur-
geons now use a gamma probe in the operating suite to
help localize the hyperfunctioning gland(s).
Post-operative recurrence rates are in the range of
5–10%. Common reasons for surgical failure include: (1)
an abnormal location of the tumor within the neck or
mediastinum, (2) failure to recognize hyperplasia, or (3)
an undiscovered fifth gland. Re-exploration has an
increased morbidity and poorer success rate than the pri-
mary procedure. Thus preoperative imaging is particu-
larly valuable in this group of patients.
Preoperative Noninvasive Imaging
Various imaging techniques have been used to preopera-
tively localize hyperfunctioning parathyroid glands. The
accuracy of ultrasonography, computed tomography, and
magnetic resonance imaging are inferior to present-day
methods of performing scintigraphy. The other imaging
modalities are particularly insensitive for detecting
ectopic and mediastinal glands. Numerous different
scintigraphic protocols have been used over the years.
Tl-201/Tc 99m pertechnetate imaging has long been
abandoned in favor of Tc-99m sestamibi protocols.
Radiopharmaceuticals
Selenium-75 selenomethionine was the first radio-
pharmaceutical used for parathyroid scintigraphy in
1965. This amino acid analog of methionine is incorpo-
rated into areas of protein synthesis. Image quality was
poor and accuracy was suboptimal.
The thallium-201/Tc-99m pertechnetate subtraction
technique was used in the 1980s. Tl-201 is a nonspecific
Endocrine System 103
tumor imaging agent taken up by both benign and malig-
nant tumors, including hyperfunctioning parathyroid
glands. Being a blood flow agent, it is also avidly taken up
by normal thyroid. Tc-99m pertechnetate is only extracted
by thyroid tissue.
With commonly used methodology,Tl-201 was adminis-
tered first and a neck image acquired on computer. Then
Tc-99m pertechnetate was administered and another
image acquired. The Tc-99m activity was then computer
subtracted from the Tl-201 (Fig. 5-30). This method has
been abandoned because of the poor imaging characteris-
tics of thallium-201, difficulties associated with patient
motion and computer image registration, variable
reported accuracy,and the advent of Tc-99m sestamibi.
Technetium-99m Sestamibi and Tc-99m Tetrofosmin
Mechanism of Uptake
Tc-99m sestamibi (Cardiolite), like Tl-201, is a cardiac
imaging agent. Sestamibi is a lipophilic cation and mem-
ber of the isonitrile family (hexakis 2-methoxyisobutyl
isonitrile). The mechanism of uptake is probably related
to high cellularity and tumor vascularity. It localizes
within the cytoplasm and mitochondria of the cell
because of electrical potentials. The large number of
mitochondria oxyphil cells of parathyroid adenomas may
be responsible for its avid uptake and slow release com-
pared to surrounding tissue. Tc-99m tetrofosmin is simi-
lar to Tc-99m sestamibi in its mechanism of uptake;
however, data are more limited.
Pharmacokinetics
Tc-99m sestamibi has higher uptake than Tl-201, superior
image quality, and lower radiation dosimetry. Peak accu-
mulation occurs at 3–5 minutes with a half-time clear-
ance of approximately 60 minutes. In addition, sestamibi
has interesting washout characteristics. Tc-99m ses-
tamibi usually clears slower from hyperfunctioning
parathyroid tissue than adjacent thyroid tissue. This is
the rationale for early (10 min) and late (2 hour) imaging.
By delayed imaging, the thyroid has usually cleared, leav-
ing uptake in the hyperfunctioning parathyroid tissue
(Fig. 5-31).
Methodology
Tc-99m sestamibi, 20–25 mCi, is injected intravenously.
The study is commonly performed in two phases, early
planar imaging at 10 minutes and late planar imaging at
2 hours after injection (Box 5-18). This permits diagno-
sis by observing the differential washout from the thy-
roid and hyperfunctioning parathyroid. Some centers use
pinhole imaging of the neck and parallel hole imaging of
the chest in order not to miss a mediastinal adenoma;oth-
ers use only one large field of view using a high-resolu-
tion parallel hole collimator. Dual-isotope techniques
using I-123 thyroid subtraction are used at some centers.
Single-photon emission computed tomography (SPECT)
is increasingly being performed. Because the length of
104 NUCLEAR MEDICINE: THE REQUISITES

Figure 5-30 Thallium-21/Tc-99m pertechnetate parathyroid subtraction scan. Patient has

hypercalcemia and increased serum parathormone level. A, Tc-99m pertechnetate scintigraphy in
a patient is essentially normal. B, Corresponding Tl-201 scintigraphy reveals an apparent area of
increased uptake adjacent to the lower pole of the right lobe. C, Subtraction of the Tc-99m
pertechnetate study from the T1-201 study confirms the presence of the parathyroid adenoma.

time for acquisition is long, only one imaging period is
commonly used (Fig. 5-32).
Image Interpretation
Initial images at 10–15 minutes after injection show
prominent thyroid uptake. Many times, focal uptake is
seen in a parathyroid adenoma at this time point. On
delayed imaging, much of the thyroid uptake has washed
out and the typical finding of a hyperfunctioning
parathyroid gland is a focus of residual activity in the
neck with a high target-to-background ratio. On occa-
sion, multiple adenomas or hyperplastic glands may be
seen (Fig. 5-33).

Figure 5-31 Tc-99m sestamibi parathyroid scan. Patient has hypercalcemia and increased PTH.
A, Early imaging at 15 minutes with Tc-99m sestamibi reveals somewhat asymmetrical activity in the
region of the thyroid gland. B, Delayed imaging at 2 hours demonstrates washout of thyroid activity
and a parathyroid adenoma.
In a minority of patients, the washout rate of the thy-
roid and hyperfunctioning parathyroid will be similar.
However, often in these cases, the parathyroid adenoma
can be seen as a distinct focus with a background of thy-
roid activity. Although adenomas are most commonly
detected contiguous to the thyroid or occasionally
intrathyroidal, they may be ectopic, anywhere from high
in the neck down to the mediastinum. The advantage of
SPECT is increased sensitivity (although direct compari-
son data is limited) and better anatomical localization
(e.g., defining a retrotracheal gland) (see Fig. 5-32).
Accuracy
The sensitivity for detection of parathyroid adenomas
larger than 300 mg in size is greater than 85–90% but is
less for smaller adenomas. The most common cause for
a false negative study is the small size of the adenoma.
The sensitivity for detection of hyperplasia is consider-
ably lower than adenoma (~50–60%). The most com-
mon cause for a false positive study is a thyroid adenoma.
ADRENAL SCINTIGRAPHY
Different radiopharmaceuticals are available for scinti-
graphic imaging of the adrenal cortex and the adrenal
medulla. Adrenocortical scintigraphy was used before
the development of CT and MRI. Nuclear imaging studies
of the adrenal cortex are not frequently performed in
Endocrine System 105
current practice but retain a limited role for assessing the
functional status of adrenocortical tissue when other
imaging studies are indeterminate. However, scinti-
graphic studies of the adrenal medulla and related tissues
have found an increasing role in contemporary practice.
Adrenocortical Scintigraphy
The most common indication for adrenocortical scintig-
raphy is hypercortisolism (Cushing’s syndrome or
Cushing’s disease). Less common indications include
hyperaldosteronism (Conn’s syndrome) and adrenal-
virilizing tumors. These hormones are derived from dif-
ferent layers of the adrenal cortex: the zona fasciculata,
zona glomerulosa, and zona reticularis, respectively.
Radiopharmaceuticals
I-131-6β-iodomethyl-19-norcholesterol (NP-59) localizes
in the adrenal cortex as a result of the transport and
receptor systems for serum cholesterol bound to low-
density lipoprotein (LDL). Factors affecting cholesterol
uptake into the adrenal also affect uptake of the radio-
pharmaceutical. Thus,elevated serum cholesterol reduces
the percent uptake. Increases in plasma adrenocorti-
cotropic hormone (ACTH) result in increased radiocho-
lesterol uptake. The radiopharmaceutical is stored in
adrenocortical cells and is esterified but not incorporated
into adrenal hormones.
106 NUCLEAR MEDICINE: THE REQUISITES
Box 5-18 Tc-99m Sestamibi Parathyroid
Imaging: Protocol Summary
PATIENT PREPARATION
None
RADIOPHARMACEUTICAL
20 mCi (740 MBq), intravenously
TIME OF IMAGING
Early scans at 15 minutes
Delayed scans at 2 hours
IMAGING PROCEDURE
Planar
Use a high-resolution collimator and a 20% window
centered at 140 keV.
Position the patient supine with the chin up and neck
extended.
Place markers on the chin and sternal notch.
Obtain anterior and 45-degree left and right anterior
oblique views, 300k counts per view.
SPECT IMAGING
Position patient as above.
Use a high-resolution collimator and a 20% window
centered at 140 keV.
Use dual-headed SPECT camera: 360-degree contoured
acquisition arc, 3-degree angular
Sampling increment, 15–30 sec per view, 128 × 128
matrix with 1.5 zoom, Hanning or Butterworth filter.
Reconstruct transaxial, coronal, and sagittal planes.
Reproject images at each sampling angle.
The uptake of I-131-6β-iodomethyl-19-norcholesterol
is progressive over several days after tracer administra-
tion. Background clearance is also relatively slow, and for
routine or baseline studies, imaging is typically per-
formed several days after tracer injection. Background tis-
sues demonstrating significant localization include the
liver, colon, and gallbladder.
Patients should be pretreated for at least 1 day with
iodine (e.g., SSKI, 1 drop three times a day or equivalent)
to block uptake of free radioiodine in the thyroid. This is
continued for 7 days (Box 5-19).
I-131-6β-iodomethyl-19-norcholesterol is usually given
in a dose of 1 mCi/1.7 m2 of body surface area. The dose
is administered intravenously over 1–2 minutes. For
hypercortisolism, imaging is acquired at 48 hours after
radiopharmaceutical injection. For other indications,
imaging is initiated at 4–5 days because the lower level of
uptake requires more time for background clearance and
because of the suppression protocol described later.
A large field-of-view gamma scintillation camera with
a high-energy parallel-hole collimator is used and a 20%
window is centered at 364 keV. A standard imaging time
of 20 minutes per view is used. The most important
view is posterior and includes both adrenal glands.
Anterior views may be helpful to assess adrenal asymme-
try. Lateral views can help differentiate normal gallblad-
der uptake from activity in the right adrenal gland.
Suppression Studies
In patients with hyperfunctioning of the zona glomeru-
losa (hyperaldosteronism) or the zona reticularis (hyper-
androgenism), it is desirable to suppress ACTH secretion
and thus uptake of radiocholesterol in the zona fascicu-
lata site of cortisol production. Without suppression,
normal high uptake in the zona fasciculata would make
interpretation of uptake by the zona glomerulosa or
reticularis quite difficult. Suppression is accomplished
by administering dexamethasone, 4 mg per day (2 mg
bid) for 7 days before radiopharmaceutical administra-
tion and continuing until imaging is completed.
Normal Adrenocortical Scintigram
In normal subjects, radiotracer uptake in the adrenal cor-
tex increases over the first 2 days after injection.
Background activity is still relatively high at this time,
especially in the liver, and imaging may be delayed until
day 4 or 5.
The two adrenal glands are not anatomically symmet-
rical and often have different scintigraphic appearances.
The right adrenal typically sits at the superior pole of the
right kidney and is slightly cephalad to the left adrenal
gland. The right adrenal gland appears round and in
most subjects is slightly more intense than the left
because of its more posterior location in the body and
less soft tissue attenuation. Liver activity is also superim-
posed. The left adrenal gland typically lies at the antero-
medial border of the left kidney and may extend
inferiorly to the renal hilum. It appears more caudad and
has an oval rather than a round configuration. The left
adrenal gland frequently appears less intense because of
its more anterior location and the lack of additive back-
ground liver activity.
Cushing’s Syndrome
The scintigraphic pattern of uptake depends on the etiol-
ogy of hypercortisolism (Fig. 5-34). In Cushing’s disease,
caused by a pituitary adenoma with increased production
of ACTH, there is bilateral early visualization of the
adrenal glands (Fig. 5-35). Ectopic ACTH production
would produce a similar scintigraphic pattern, although
it is not a common indication for scintigraphy. Unilateral
visualization is classically seen in patients with glucocor-
ticoid-producing adrenal adenomas (Fig. 5-36). The
autonomous production of cortisol in the adenoma feeds
back to shut off pituitary ACTH secretion and thereby
shuts off uptake in the contralateral adrenal gland
 Endocrine System 107

Figure 5-32 SPECT sestamibi parathyroid scan. Preoperative hyperparathyroidism. Sequential

coronal (above) and transverse (below). Posterior inferior parathyroid adenoma.

(Fig. 5-37). Adrenal macronodular hyperplasia is an
uncommon form of hypercorticalism in which there is
autonomous function of both adrenal glands and bilateral
uptake scintigraphically, although often somewhat asym-
metrically.
Nonvisualization of both adrenal glands in patients with
Cushing’s syndrome indicates adrenal carcinoma. The
tumors can be quite large and are often first manifested
clinically with signs and symptoms of hormone excess.
However, the function per gram of tumor tissue is typically
low and tracer uptake is insufficient to visualize the tumor.
The contralateral adrenal gland is not visualized because
the excessive cortisol production shuts down pituitary
ACTH secretion. Biochemical proof of hypercortisolism
and CT demonstration of a large lesion in the adrenal are
considered sufficient evidence and scintigraphy is not
needed. However, if CT findings are negative or equivocal,
adrenocortical scintigraphy can be helpful.
One use of adrenocortical scintigraphy in patients with
Cushing’s syndrome, even in the era of MRI and CT, is in
the detection of postsurgical adrenal remnants. These
remnants may cause recurrent disease and be difficult
108 NUCLEAR MEDICINE: THE REQUISITES

Figure 5-33 Tertiary hyperparathyroidism with four glands detected.Tc-99m sestamibi scan
in patient with renal failure and elevated calcium. SPECT coronals with four abnormal foci, the left
superior parathyroid gland being the largest, but also the left inferior, right superior and right
inferior glands detected. Confirmation of scintigraphic findings at surgery.

are typically small and CT or MRI often is not diagnostic.

Box 5-19 Iodine Available for Thyroid Aldosterone is produced in the zona glomerulosa of the
Radiation Protection adrenal cortex. This hormone does not affect the pitu-
itary–ACTH feedback loop. Dexamethasone suppression
Lugol’s solution 6.3 mg/drop scan is necessary for scintigraphic evaluation of patients
Supersaturated potassium 38 mg/drop with aldosteronism.Normal uptake in the zona fasciculata
iodide (SSKI) can obscure asymmetry caused by small nodules and ade-
Quadrinol 145 mg/tab nomas in the zona glomerulosa. Scintigraphy is performed
Potassium iodide capsules 130 mg/capsule over several days.Early (<5 days) unilateral “breakthrough”
indicates aldosteronoma. Bilateral, delayed breakthrough
typically indicates hyperplasia.
to localize in a surgically altered anatomy. They can be
detected by adrenocortical scintigraphy. Androgen Excess
The adrenal glands may be the source of excessive pro-
Hyperaldosteronism duction of androgen. Scintigraphic patterns are similar to
The principal clinical question in aldosteronism is the dis- those found in aldosteronism. That is, patients with bilat-
tinction of adenoma from hyperplasia. Aldosteronomas eral hyperplasia demonstrate bilateral breakthrough on

Symmetric Bilateral hyperplasia

Bilateral
visualization Asymmetric Bilateral hyperplasia
(some asymmetry is
common)
Bilateral hyperplasia
with associated
Figure 5-34 Cushing’s syndrome: NP-59
unilateral adenoma
(common) diagnostic patterns. I-131 norcholesterol (NP-59)
can aid in the differential diagnosis. Diagnostic
Adrenal remnants patterns in patients with biochemically proven
Posterior after adrenalectomy Cushing’s syndrome include unilateral uptake for
adrenal Unilateral
visualization Adenoma adenomas, bilateral uptake for pituitary etiology,
scintigram
and no uptake for adrenocortical carcinoma.

Bilateral
nonvisualization Carcinoma

Drug therapy
 Endocrine System 109

Figure 5-35 Adrenal adenoma. Adrenocortical scintigraphy in the posterior view reveals
unilateral uptake in the left adrenal gland in a patient with adrenal adenoma. The anterior view
shows uptake in the adenoma as well as in the gallbladder that was confirmed by a lateral view.

Hypothalamus dexamethasone suppression scans and adenomas are

characterized by marked scintigraphic asymmetry.

CRF (⫹) Incidentalomas

I-131 MIBG has been used as an aid in determining the
etiology of a nodule found incidentally (e.g., on CT).
Uptake in the nodule signifies a functioning nodule and
ACTH (⫹)
not metastatic disease.
Pituitary
(⫺)
Adrenomedullary Scintigraphy
Zona fasciculata
Cortisol Adrenomedullary scintigraphy has proven useful in the
management of patients with functional adrenergic
Androgens Zona reticularis
tumors, including paragangliomas, neuroblastomas, and
Medulla pheochromocytomas. Pheochromocytomas are paragan-
Aldosterone
Zona glomerulosa gliomas that arise in the adrenal medulla. Paragangliomas
are associated with a number of important familial syn-
Adrenal
dromes, including multiple endocrine neoplasia (MEN)
Figure 5-36 Pituitary–adrenal feedback loop. Function in the
zona fasciculata is stimulated by adrenocorticotropic hormone type IIA (medullary carcinoma of the thyroid, pheochro-
(ACTH). Pituitary secretion of ACTH decreases as circulating levels mocytoma, and hyperparathyroidism) and MEN type IIB
of cortisol increase. CRF, corticotropin-releasing factor. (medullary carcinoma of the thyroid,pheochromocytoma,

Figure 5-37 I-131 MIBG localizes pheochromocytoma. Adrenomedullary scintigraphy reveals

unilaterally increased uptake in the region of the left adrenal due to pheochromocytoma. The
patient had elevated nor- and metanephrines.
110 NUCLEAR MEDICINE: THE REQUISITES

and ganglioneuromas), von Hippel-Lindau disease, and

neurofibromatosis. Table 5-3 Drugs Known or Expected to Reduce
MIBG Uptake
Radiopharmaceuticals
Scintigraphic studies of the adrenergic nervous system KNOWN (PATIENT MUST BE OFF PRIOR TO SCANNING)
became possible with meta-iodo-benzyl-guanidine Antihypertensive/Cardiovascular
(MIBG), a norepinephrine analogue. Localization is via Labetalol, reserpine
the type I, energy-dependent, active amine transport Calcium-channel blockers (diltiazem, nifedipine, verapamil)
mechanism. The tracer is taken up and localized in cyto- Tricyclic Antidepressants
plasmic storage vesicles in presynaptic adrenergic
Amitriptyline and derivatives
nerves. In addition to the uptake in the adrenal medulla Imipramine and derivatives
and other adrenergic and neuroblastic tumor tissues, the Doxepin, amoxapine, loxapine (antipsychotic agent)
tracer localizes avidly in other organs with rich adrener-
Sympathomimetics
gic innervation, including the heart, salivary glands, and
spleen. Both I-131 and I-123 have been used as radiola- Phenylephrine, phenylpropanolamine, pseudoephedrine,
ephedrine
bels. I-123 has the advantage of a lower radiation dose to
the patient, whereas I-131 allows for delayed imaging. Cocaine

EXPECTED (MAY INTERFERE BUT NEED NOT BE CEASED

Technique PRIOR TO SCANNING)
MIBG is taken up rapidly by adrenergic tissues. To Antihypertensive/Cardiovascular
achieve a sufficient target-to-background ratio, imaging is Adrenergic neurone blockers (bethanidine, debrisoquine,
typically delayed for 24–48 hours, usually longer. bretylium, guanethidine)
For studies with I-131 or I-123 MIBG, patients are
“Atypical” Antidepressants
given a blocking dose of saturated solution of potassium
iodide (SSKI). The usual adult dose of I-131 MIBG is Maprotiline, Trazodone
0.5 mCi/1.7 m2. The tracer is administered intravenously POSSIBLE
over 15–30 seconds. When MIBG is radiolabeled with
Antipsychotics (major tranquilizers)
I-123, up to 10 mCi/m2 can be administered with the
same radiation dose to the patient as from 0.5 mCi/m2 of Phenothiazines (chlorpromazine, triflupromazine,
promethazine, etc.)
I-131 MIBG. Thioxanthenes (chlorprothixene, thiothixene)
Initial images with I-131 MIBG may be obtained at Butyrophenones (droperidol, haloperidol, pimozide)
24 hours. The optimal imaging time is 48 hours after
Sympathomimetics
injection, and 72-hour imaging is possible. A wide field-
of-view gamma scintillation camera equipped with Amphetamine and related compounds
Beta-sympathomimetics: albuterol, isoetharine,
a high-energy parallel-hole collimator is used. Imaging is isoproterenol, metaproterenol, terbutaline
typically for 20 minutes. The views obtained are deter- Dobutamine, dopamine, metaraminol, tetrabenazine
mined by the clinical condition under evaluation. For
NO EFFECT
pheochromocytoma, the posterior view of the mid-
abdomen with the region of the adrenal glands is most Antihypertensive/Cardiovascular
important. Additional images from the pelvis to the base Alpha-blockers (clonidine, phenoxybenzamine,
of the skull are indicated to detect extra-adrenal pheochro- phentolamine, prazosin)
Alpha-methyldopa
mocytoma (paraganglioma) and neuroblastomas. Angiotensin converting enzyme inhibitors (captopril,
With I-123 MIBG, initial images may be obtained at enalapril)
4 hours, although images at 24 hours are superior. SPECT Beta-blockers (does not include labetalol)
is feasible with I-123 MIBG. The significantly lower radia- Digitalis glycosides, diuretics
tion dose and superior image quality is a major advantage Analgesics
for children with neuroblastoma. Major (morphine and other opioids), minor (aspirin,
acetaminophen)
Precautions
Hypnotics, minor tranquilizers
A number of drugs interfere with MIBG uptake and
a drug history should be obtained before imaging.
Interfering drugs include tricyclic antidepressants, reser-
pine, guanethidine, certain antipsychotics, cocaine, and
the alpha- and beta-blocker labetalol (Table 5-3).
 Endocrine System 111

Normal MIBG Scintigraphy

With the usual doses used for I-131 MIBG imaging, only
faint visualization of the normal adrenal medulla is
achieved in 10% of patients. The normal adrenal medulla
is visualized more frequently with I-123 MIBG. Early
images reveal activity in the spleen, heart, salivary glands,
and liver. These areas clear with time. Some bladder
activity may be visualized because of free radioiodine.
The colon is also seen transiently in 20% of cases.

Clinical Applications
The greatest clinical experience with MIBG is in the eval-
uation of patients with suspected pheochromocytoma.
However, it has an important role in the evaluation and
follow-up of patients with neuroblastoma.
Pheochromocytoma
The characteristic appearance is unilateral focal uptake in
the tumor (Fig. 5-38). Sensitivity for detection of
pheochromocytoma is approximately 90%, with speci-
ficity being greater than 95%. In approximately 10% of
patients, pheochromocytoma is bilateral. In 10–20%
of patients, the tumors are extra-adrenal and are referred
to as paragangliomas. Pheochromocytomas are increas-
ingly seen with other neuroectodermal disorders, includ-
ing neurofibromatosis, tuberous sclerosis, Carney’s
syndrome,and von Hippel-Lindau disease.Paragangliomas
may be found from the bladder up to the base of the skull.
Scintigraphy with MIBG is not a screening procedure
for pheochromocytoma and should be applied only after
biochemical tests suggest the diagnosis. Many centers
first use CT to evaluate the adrenal glands. If an adrenal
mass is demonstrated, the diagnosis is inferred and fur- Figure 5-38 I-131 MIBG: neuroblastoma follow-up. Two-year-
ther workup before surgery is often unnecessary. MIBG is old with stage 4 neuroblastoma status post left retroperitoneal
particularly helpful in surveying the entire body for mass resection, bone marrow transplant, and monoclonal antibody
therapy. The 48-hour I-131 scan shows a large left upper intra-
extraadrenal tumors and metastatic disease. abdominal mass seen with greatest uptake at the periphery. Bone
Adrenomedullary hyperplasia develops in patients metastases in the skull and right femur.
with MEN type IIA. This condition is difficult to diagnose
with CT or MRI. MIBG scintigraphy is uniquely suited to
detect medullary hyperplasia and has been used to assist the bone marrow. The combination of the two results in
decision making for timing of surgery. the highest sensitivity for detection.
Neuroblastoma The sensitivity for MIBG for neuroblastoma is greater
This malignant tumor of neural crest origin occurs in than 90%, with a high degree of specificity. MIBG is used
young children, usually less than 4 years of age. Seventy for staging (Fig. 5-39), detecting metastatic disease, and
percent of tumors originate in the retroperitoneal following the patient’s response to therapy.Whole body
region, either from the adrenal or the abdominal sympa- scanning is routine for imaging patients with this disease.
thetic chain. However, 20% occur in the chest, deriving The higher count rate, better image quality, and lower
from the thoracic sympathetic chain. More than 90% of dosimetry of I-123 MIBG makes it preferable over I-131
neuroblastomas produce catecholamines. MIBG in children.
Bone scans have traditionally been used to detect Other tumors demonstrating uptake of MIBG include
metastases. However, a common location for metastases carcinoid tumors and medullary carcinoma of the thy-
is in the metaphyseal region, which can be hard to detect roid. However, the sensitivity for tumor detection is
because of the high uptake in growth plates. MIBG has lower than for neuroblastoma or pheochromocytoma.
superior sensitivity for detection of metastases compared The high uptake of MIBG in these tumors has led inves-
to the bone scan because these tumors initially involve tigators to attempt therapy with I-131 MIBG. Therapeutic
112 NUCLEAR MEDICINE: THE REQUISITES
Figure 5-39 I-123 MIBG: neuroblastoma follow-up. A 4-year-
old with metastatic neuroblastoma since age 2. Postresection of
right adrenal mass and adenopathy, chemotherapy, radiation, stem
cell transplant. Large intra-abdominal tumor mass. I-123 4-hour
scan only because family refused 24-hour imaging. Considerable
background uptake in lungs, abdomen, and soft tissue. Focal
uptake in the proximal and distal femur.An intra-abdominal tumor
mass was seen on SPECT (not shown). Nephrostomy tube in place
for urinary obstruction.
applications are still investigational and restricted largely to
patients in whom prior conventional therapies have failed.
SUGGESTED READING
Thyroid Imaging and Function Studies
Chapman EM: History of the discovery and early use of radioac-
tive iodine. JAMA 250: 2042-2044, 1983.
Filesi M, Signore A,Ventroni G, et al: Role of initial iodine-131
whole-body scan and serum thyroglobulin in differentiated thy-
roid carcinoma metastases. J Nucl Med 39: 1542-1546, 1998.
Freitas JE: Changing concepts in the management of thyroid
cancer. In Nuclear medicine annual. Leonard M. Freeman, Ed.
Philadelphia, Lippincott Williams & Wilkins, 2003, pp. 101-130.
Grunwald F, Schomburg A, Bender H, et al: Fluorine-18–fluo-
rodeoxyglucose positron emission tomography in the follow-up of
differentiated thyroid cancer. Eur J Nucl Med 23:312-319,1996.
Gulec MG, Rubello D, Boni G, et al: Preoperative localization and
radioguided parathyroid surgery. J Nucl Med 44: 1443-1458,
2003.
Intenzo CM, dePapp AE, Jabbour S, et al: Scintigraphic manifesta-
tions of thyrotoxicosis. Radiographics 23: 857-869, 2003.
Intenzo CM, Capuzzi DM, Jabbour S, et al: Scintigraphic features
of autoimmune thyroiditis. Radiographics 21: 957-964, 2001.
Park HM, Perkins OW, Edmondson JW, et al: Influence of diag-
nostic radioiodines on the uptake of ablative dose of iodine-
131.Thyroid 4: 49-54, 1994.
Shankar LK, Yamamoto AJ, Alavi A, Mandel SJ: Comparison of
I-123 scintigraphy at 5 and 24 hours in patients with differenti-
ated thyroid cancer. J Nucl Med 43: 72-76, 2002.
Smith JR and Oates E: Radionuclide imaging of the thyroid
gland: patterns, pearls, and pitfalls. Clin Nucl Med 29: 181-193,
2004.
Uematsu H, Sadato N, Ohtsubo T, et al: Fluorine-18-fluo-
rodeoxyglucose PET versus thallium-201 scintigraphy evalua-
tion of thyroid tumors. J Nucl Med 39: 453-459, 1998.
Adrenal Scintigraphy
Gelfand MJ: Meta-iodobenzylguanidine in children. Semin Nucl
Med 23: 231-242, 1993.
Gross MD, Shapiro B, Frances IR, et al: Scintigraphic evaluation
of clinically silent adrenal masses using adrenocortical scintigra-
phy. J Nucl Med 35: 1145-1152, 1994.
Hay RV, Shapiro B, Gross MD: Scintigraphic imaging of the
adrenals and neuroectodermal tumors. In Nuclear medicine.
Henkin RE, Boles MA, Dillehay GL, et al, Eds. St Louis, Mosby,
1996.
Sisson JC: Scintigraphic localization of pheochromocytomas.
N Engl J Med 305: 12-17, 1981.
Parathyroid Scintigraphy
Civelek AC,Ozalp E,Donovan P,Udelsman R:Prospective evalua-
tion of delayed technetium-99m sestamibi SPECT scintigraphy
for preoperative localization of primary hyperparathyroidism.
Surgery 131: 149-157, 2002.
Perez-Monte JE, Brown ML, Shah AN, et al: Parathyroid adeno-
mas: accurate detection and localization with Tc-99m sestamibi
SPECT. Radiology 201: 85-91, 1996.
Taillefer R, Boucher Y, Potvin C, Lambert R: Detection and local-
ization of parathyroid adenomas in patients with hyperparathy-
roidism using a single radionuclide imaging procedure with
technetium-99m sestamibi (double-phase study). J Nucl Med
33:1801-1807, 1992.
 CHAPTER Skeletal Scintigraphy

6
Bone Scan Introduction Shin Splints
Radiopharmaceuticals Rhabdomyolysis
History Heterotopic Bone Formation
Preparation Bone Infarction and Osteonecrosis
Uptake and Pharmacokinetics Legg-Calvé-Perthes Disease
Dosimetry Steroid-Induced Osteonecrosis
Imaging Protocol Sickle Cell Anemia
Normal and Altered Distribution Osteomyelitis
Clinical Uses of Skeletal Scintigraphy Scintigraphic Findings of Osteomyelitis
Metastatic Disease Prosthesis Evaluation
Metastatic Disease in Specific Tumors Bone Marrow Scintigraphy
Prostate Carcinoma Bone Mineral Measurement
Breast Carcinoma
Lung Carcinoma
Scintigraphic Patterns in Metastatic Disease BONE SCAN INTRODUCTION
Solitary Lesions
Superscan The skeleton is an active, constantly changing organ. It is
Flare Phenomenon made up of inorganic calcium hydroxyapatite crystal,
Cold Lesions Ca10(PO4)6(OH)2, and an organic matrix of collagen and
Extraskeletal Uptake in Soft Tissues blood vessels. Bone responds to injury and disease with
Imaging Findings in Specific Tumors increased turnover and attempts at self-repair.This physi-
Breast Carcinoma ologic process can be imaged with a radiotracer that
Lung Carcinoma localizes to areas of bone formation.
Neuroblastoma Bone scans have used the technetium-99m labeled
Other Tumors of Epithelial Origin diphosphonates such as Tc-99m methylene diphospho-
Primary Tumors nate (Tc-99m MDP) for decades to perform skeletal imag-
Malignant Tumors ing. The bone scan is a versatile tool that can image
Benign Bone Tumors malignant and benign processes.The bone scan is highly
Metabolic Bone Disease sensitive for disease, is readily available, and can image
Hyperparathyroidism the entire skeleton at reasonable cost. Therefore, skeletal
Osteoporosis scintigraphy (Fig. 6-1) remains popular despite techno-
Paget’s Disease logical advances in magnetic resonance imaging (MRI ),
Skeletal Trauma computed tomography (CT ), and positron emission
Detection of Fractures tomography (PET).
Iatrogenic Trauma The major drawback of this exam is its low specificity.
Child Abuse Numerous benign processes (e.g., arthritis) cause in-
Complex Regional Pain Syndrome creased radiotracer uptake by increasing blood flow and
Stress Fractures osteogenic activity. Often the location and patterns of

113
114 NUCLEAR MEDICINE: THE REQUISITES

These women ingested the bone seeker,Radium-226,when

licking their brushes to tip them. Since then, numerous
bone-seeking agents have been studied and abandoned.
This includes Phosphorus-32,radioisotopes of calcium,sev-
eral rare earth elements, and isotopes of gallium, barium,
samarium, and strontium. Gallium-67 is still used in tumor
and infection imaging.Fluorine-18 (F-18) is an analog of the
hydroxyl ion found in calcium hydroxyapatite and avidly
localizes to bone. It was the agent of choice for skeletal
scintigraphy until the advent of the Tc-99m phosphonates
in the 1970s.
An ideal radiopharmaceutical for skeletal scintigraphy
must be inexpensive, remain stable, rapidly localize to
bone, quickly clear from the background soft tissues, and
have favorable imaging and dosimetry characteristics.
These parameters were essentially met in the 1970s
when technetium-99m, already desirable for gamma cam-
era imaging studies, was combined with members of the
phosphate family.
These radiopharmaceuticals are classified by the type
of phosphate bond. The first of these agents, pyrophos-
phates and then the longer-chain polyphosphates, were
soon replaced by the diphosphonates (Fig. 6-2). The
diphosphonates are more stable in the body and have
better background clearance than pyrophosphates or
polyphosphates. The diphosphonate agents include
Tc-99m hydroxyethylidene diphosphonate (Tc-99m
HEDP),Tc-99m hydroxymethylene diphosphonate (Tc-
99m HMDP or HDP), and Tc-99m methylene diphospho-
nate (Tc-99m MDP). The ability of each diphosphonate
to detect lesions has been studied. Although some differ-
ences are present,Tc-99m MDP and Tc-99m HDP are both
excellent agents.

Preparation
Tc-99m MDP can be prepared from a simple kit.
Technetium-99m, in the form of sodium pertechnetate
Figure 6-1 Normal Tc-99m methylene diphosphonate (MDP) (NaTcO4), is obtained from a molybdenum-99 generator
whole body bone scan. A high level of anatomic detail can be
visualized. Some areas of increased uptake are normally seen in the and injected into a vial containing methylene diphospho-
adult including activity in the joints. nate, stabilizers, and stannous ion. Stannous tin acts as
a reducing agent which allows the technetium-99m to
form a chelate bond with the methylene diphosphonate
carrier molecule.
abnormalities can guide interpretation. However, it is
essential to know the patient’s history, understand when
radiographic correlation is necessary, and know how to O O O R1 O
use it.
⫺O P O P O⫺ ⫺O P C P O⫺

Radiopharmaceuticals O⫺ O⫺ O⫺ R2 O⫺
History
It has long been known that certain radioisotopes localize Pyrophosphate Diphosphonate
to bone.In the 1920s,devastating cancers were reported in Figure 6-2 Chemical structures of pyrophosphate and
women who painted luminescent watch dials with radium. diphosphonate.

Incomplete labeling may occur if air is introduced into
the vial causing hydrolysis of the stannous ion (from Sn II
into Sn IV). If not enough stannous ion is available to
reduce the technetium ion, free technetium pertechne-
tate (“free tech”) will result, causing image degradation
as a result of increased soft tissue uptake distribution and
uptake in thyroid, stomach, and salivary glands (Fig. 6-3).
Occasionally, the excess Sn II may form a partly colloidal
radiopharmaceutical, which can accumulate in the
reticuloendothelial system of organs such as the liver.
Tc-99m MDP should be used within 2–3 hours of prepa-
ration or radiopharmaceutical breakdown may also yield
technetium pertechnetate.
Uptake and Pharmacokinetics
After intravenous injection,Tc-99m MDP rapidly distrib-
utes into the extracellular fluid and is quickly taken up
into the bone. Tc-99m MDP accumulates primarily in
relation to osteogenic activity levels, although the
amount of blood flow plays a part. Activity is much
higher in areas of active bone formation compared with
mature bone. Tc-99m MDP binding occurs by chemoad-
sorption in the hydroxyapatite mineral component of
the osseous matrix. Uptake in areas of amorphous cal-
Figure 6-3 Free pertechnetate in the radiopharmaceutical
preparation. This has resulted in uptake in stomach, thyroid gland,
and salivary glands. Salivary activity has entered the oropharynx
on this 3-hour delayed image.
Skeletal Scintigraphy 115
cium phosphate may account for Tc-99m MDP uptake
in sites outside the bone, such as dystrophic soft tissue
ossification.
Approximately 50% of the dose is localized to the
bone with the remainder excreted by the kidneys.
Although peak bone uptake occurs approximately
1 hour after injection, highest target-to-background ratios
are seen after 6–12 hours. This must be balanced with
the relatively short 6-hour half-life of Tc-99m and patient
convenience. Therefore, images are typically taken 2–4
hours after injection. Serum radiotracer levels at this time
are down to 3–5% of the injected dose in patients with
normal renal function. It should be noted that the half-life
of Tc-99m effectively limits imaging to within approxi-
mately 24 hours of injection.
Decreased localization is seen in areas of reduced or
absent blood flow or infarction. Diminished uptake is
also seen in areas of severe destruction that can occur in
some very aggressive metastasis (Fig. 6-4). Photon defi-
cient areas are sometimes referred to as “cold.”
Dosimetry
Estimates of absorbed radiation doses are listed in
Table 6-1. The radiation dose to the bladder wall,ovaries,
and testes depends on the frequency of voiding. The
dosimetry provided assumes a 2-hour voiding cycle.
Significantly higher doses result if voiding is infrequent.
Radiopharmaceuticals are administered to pregnant
women only if clearly needed on a risk-versus-benefit
basis. Tc-99m is excreted in breast milk so breastfeeding
should be stopped for 24 hours.
Imaging Protocol
An example protocol is listed in Box 6-1. There are sev-
eral modifications possible in skeletal scintigraphy proto-
cols.It must first be determined if a three-phase bone scan
is needed to assess blood flow and soft tissue activity for
Figure 6-4 Complete destruction of the L1 vertebral body with
corresponding photon-deficient (or cold) lesion.
116 NUCLEAR MEDICINE: THE REQUISITES

cases of infection and trauma. The injection site should

Table 6-1 Radiation Absorbed Dosimetry from be chosen to avoid any suspected pathology. Also, if
Technetium-99m medronate (Tc-99m MDP) comparison with the opposite hand may be needed at
any time, injection in a site such as the foot should be
Radiation Dose (rads/mCi) considered.
Although many diagnostic questions can be answered
Organ (cGy/37 MBq)
with routine delayed imaging, the three-phase bone scan
Skeleton 0.035 is helpful in addressing several problems. The most fre-
Marrow 0.0280 quent indication for three-phase imaging is to assess for
Kidneys 0.04 possible osteomyelitis. However, it is also beneficial in
Bladder 0.13 the evaluation of a painful hip prosthesis, trauma, bone
Testes 0.008
Ovaries 0.008
graft status, and reflex sympathetic dystrophy. The tech-
Whole body 0.0065 nique for dynamic scanning is summarized in Box 6-2.
If dynamic three-phase scanning is to be performed,
the area in question is positioned under the camera for
the injection. A 20 mCi (740 MBq) bolus of Tc-99m MDP
is injected intravenously. The first phase consists of
Box 6-1 Skeletal Scintigraphy: Protocol serial 2–5 second dynamic images acquired for 60 sec-
Summary for Whole Body onds. Then blood pool or soft tissue second-phase
Survey and SPECT images are obtained of this region. Additional blood
pool images can be taken in any area of secondary inter-
PATIENT PREPARATION AND FOLLOW-UP est, such as in patients with arthritis or multiple stress
Patient should be well hydrated injuries. Delayed images constitute the third phase of
Patient should void immediately before study and should a three-phase bone scan. The delayed images can be
void frequently after procedure (reduces radiation done without flow images for routine studies, such as the
dose to bladder wall) assessment of metastatic disease.
Patient should remove metal objects (jewelry, coins, After injection, the patient should be instructed to
keys) before imaging drink several cups of fluid to improve soft tissue clear-
ance and to void frequently to decrease radiation dose to
DOSAGE AND ROUTE OF ADMINISTRATION
the critical organ, the bladder. The patient returns 2–4
20 mCi (740 MBq) technetium-99m diphosphonate hours later for delayed images and voids immediately
adult dose (standard)
Intravenous injection (site selected to avoid known or
suspected pathological condition)
Adjust dosage for pediatric patients (Webster’s rule or Box 6-2 Three-Phase Skeletal
weight adjusted; Minimum 74 MBq [2 mCi]) Scintigraphy: Protocol Summary
TIME OF IMAGING
Begin imaging 2–4 hr after tracer administration RADIOPHARMACEUTICAL DOSAGE AND ROUTE OF
ADMINISTRATION
PROCEDURE Standard tracer and dosage are used and given as a bolus
Anterior and posterior views of the entire skeleton injection
Obtain a minimum of 1000k counts per view for “whole
PROCEDURE
body”imaging systems
Obtain 300k–500k counts per image if multiple spot The gamma camera is positioned before
views are used radiopharmaceutical administration immediately over
Use the highest resolution collimator that permits the site of the suspected pathological condition
imaging in a reasonable length of time
FLOW PHASE
Obtain high-count (1000k) spot views or SPECT for
more detail Dynamic 2- to 5-sec images are obtained for 60 sec after
bolus injection
SPECT*
BLOOD POOL AND TISSUE PHASE
Acquisition: contoured orbit, 128 × 128 matrix, 6-degree
intervals, 15–30 sec/stop Immediate static images for time (5 min) or counts
Reconstruction: filtered backprojection, Butterworth (300k)
filter; cut-off 0.4, power 7
SKELETAL PHASE

*Selection of SPECT acquisition and reconstruction parameters depends greatly Delayed 300k–1000k images at 2–4 hr
on available equipment and software.
 Skeletal Scintigraphy 117

before scanning. Care must be used as urinary contami-
nation frequently causes confusion or masks potential
lesion sites. All metal (such as coins or belts) should be
removed and sites of trauma or surgery noted.
Imaging is done with a low-energy, high-resolution
collimator. Delayed planar images can be obtained either
by a whole body scan or by spot views. The whole body
scan offers the advantage of a more rapid seamless cover-
age of the entire body as the camera moves over the
patient at a predetermined rate. Spot views, on the other
hand, can provide greater detail due to higher resolution
and can better define pathology by using different fixed
camera positions (Fig. 6-5). In most centers, a whole body
scan is performed with high count spot views reserved
for symptomatic areas or suspicious appearing regions.
Other modifications to consider are special views
obtained using magnified pinhole collimation and single
photon emission computed tomography (SPECT).
Magnified images with a pinhole collimator or converg-
ing collimator are commonly used in cases of
osteonecrosis of the hips and trauma to the carpal
bones. Pinhole images may also be needed in children to
better visualize the joints. Three-dimensional assess-
ment of the bones with SPECT allows for high-contrast
images that can be formatted in transaxial, sagittal, and
coronal planes. SPECT is most useful in spinal and facial
bones, where it allows better localization of uptake.
This includes determining if there is involvement of the
facets or vertebral pedicle. SPECT increases contrast of
cold and hot lesions, which improves sensitivity.
Sometimes in spondylosis, all images including MRI and
planar bone scans are normal, but SPECT is positive
(Fig. 6-6).
Normal and Altered Distribution
The normal bone scan varies dramatically with the age of
the patient. Most notably, the growing skeleton will con-
centrate radiotracer at all active growth plates (Fig. 6-7).
These areas are often the critical sites in child abuse, pri-
mary bone tumors, and osteomyelitis. Therefore, it is
essential that children are immobilized and positioned

Figure 6-5 A, Anterior and posterior whole body images of a patient with carcinoma of the
breast.Whole body images have the advantage of depicting the entire skeleton in a single view. Note
the abnormal uptake in one of the left lower posterior ribs. B, High count density spot view of left
posterior ribs from the patient in A. The location and appearance of the lesion are better delineated
in the spot view. Tracking along the rib is classic for a metastatic lesion.
118 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-6 Added value of SPECT imaging. A, A 24-year-old athlete with chronic low back pain
and negative radiographs and MRI had a near normal planar bone scan. B, SPECT images
(transaxial, sagittal, and coronal) reveal asymmetry with focal increased uptake in the right L4-5 facet
typical for spondylolysis. SPECT may be the only test that reveals the etiology of the patient’s pain.

Figure 6-7 Normal radiotracer distribution in the immature skeleton. An anterior whole body
image shows increased uptake in the growth centers. Uptake is seen in the anterior rib ends, sternal
ossification centers, and major joints.
symmetrically. By adulthood, growth plate activity dimin-
ishes and disappears.
There are numerous areas that are normal or expected
variants depending on age and history. Some bones such
as those of the sacroiliac joints normally appear to have
intense uptake. Other areas are more intense because of
configuration and proximity to the camera, such as the
iliac wings. The sternum often has residual ossification
centers and the normal sternomanubrial joint may have
increased uptake. The costochondral junction is a com-
mon site of benign uptake.
The skull is highly variable in appearance and may
show increased uptake in the frontal bone from benign
hyperostosis frontalis interna. The lateral orbits often
show normal increased and sometimes asymmetric activ-
ity where the sphenoid ridge meets the calvarium.
The normal lordotic curvature of the spine causes por-
tions of the vertebra closest to the camera to appear “hot-
ter”than other areas. The joints may be mildly asymmetric
due to use and arthritis. For example, the sternoclavicular
joints typically accumulate activity, and it has been noted
that handedness affects intensity of shoulder uptake.
The soft tissues are a critical component of scinti-
gram analysis. The patient’s body habitus must be con-
sidered as soft tissue attenuation from breast or
Skeletal Scintigraphy 119
abdomen changes intensity of the underlying bones. It
is essential to examine the soft tissues for areas of
abnormal uptake and evidence of surgery such as mas-
tectomy. It is normal to see activity in the kidneys and
bladder; absence of activity must be explained. If the
renal cortical activity is equal to or greater than the
lumbar spine, a renal abnormality or concomitant drug
therapy should be suspected (Box 6-3).
CLINICAL USES OF SKELETAL
SCINTIGRAPHY
Metastatic Disease
A significant fraction of patients with known malignancy
develop osseous metastasis (Box 6-4). Patients may pres-
ent with bone pain (50–80%) and elevated alkaline phos-
phatase (77%) but these findings are nonspecific. The
evaluation of osseous metastatic disease is the most com-
mon use of skeletal scintigraphy. Bone scan may be used
for staging, restaging, and monitoring therapy effective-
ness. The decision on which patients will need a bone
scan depends on factors such as the type and stage of
tumor, history of pain, and radiographic abnormalities.
120 NUCLEAR MEDICINE: THE REQUISITES

require a minimum mineral loss of a 50% before a lesion is

Box 6-3 Reported Causes of Bilaterally visualized. MRI is more sensitive than bone scan because
Increased and Decreased Renal signal changes in the marrow from the tumor can be visual-
Visualization on Skeletal ized directly. However, whole body MRI is not widely avail-
Scintigrams able and generally not practical at this time.
Bone scan is said to be 95% sensitive, but this sensitiv-
INCREASED UPTAKE ity depends on several factors related to tumor type.
Nephrotoxic antibiotics Areas which are predominantly osteoblastic are easily
Urinary tract obstruction seen as areas of increased activity. Lesions which are
Chemotherapy (doxorubicin, vincristine, mostly osteoclastic or lytic are more difficult to detect as
cyclophosphamide) they will appear cold or isointense. Sensitivity is highest
Nephrocalcinosis for prostate cancer, which is mainly osteoblastic. The
Hypercalcemia detection of breast and lung cancer is also very high,
Radiation nephritis although these tumors are more mixed in their lesion
Acute tubular necrosis pattern. When bone scan fails to detect the more aggres-
Thalassemia
sive or lytic lesions in lung or breast cancer, F-18 FDG
DECREASED UPTAKE PET can often detect bone marrow abnormalities. The
Renal failure sensitivity of bone scan is low for tumors that are pre-
Superscan dominantly lytic, such as multiple myeloma and renal cell
Metastatic disease carcinoma, as well as those which are contained in the
Metabolic bone disease marrow, such as lymphoma.
Paget’s disease
Osteomalacia
Hyperparathyroidism Metastatic Disease in Specific Tumors
Myelofibrosis Prostate Carcinoma
Nephrectomy Skeletal scintigraphy is very sensitive in the detection of
metastatic disease from prostate cancer. Until the introduc-
tion of the prostate specific antigen (PSA) blood test, bone
Box 6-4 Incidence of Osseous Metastasis scan was considered the most sensitive technique for
by Tumor Type detecting osseous metastasis. Serum alkaline phosphatase
measurement detects only half the cases detected by
Breast 50–85% scintigraphy.Radiographs may be normal 30% of the time.
Neuroblastoma 80% The likelihood of an abnormal scintigram correlates
Prostate 50–70% with the clinical stage, Gleason score, and PSA level. In
Hodgkin’s lymphoma 50–70% early stage I disease, scintigrams demonstrate metastasis
Ewing’s sarcoma 60% less than 5% of the time. The incidence increases to 10%
Lung 30–50% in stage II and 20% in stage III disease. In patients with
Renal cell 30–50% PSA levels less than 10 ng/ml, bone metastases are rarely
Thyroid 40% found (<1% of the time). Skeletal scintigrams are still indi-
Melanoma 30–40%
cated for symptomatic patients and for evaluation of sus-
Bladder 15–25%
Osteosarcoma 25%
picious areas seen radiographically. With increasing
Wilm’s ≤10% PSA levels, the chance of detecting metastatic disease
increases.

Over 90% of osseous metastasis distribute to the red
marrow.In adults,red marrow is found in the axial skeleton
and the proximal portions of the humeri and femurs. As
the tumor enlarges, the cortex becomes involved. The
body responds by attempts at repair. The Tc-99m MDP
binds to these regions in areas of bone deposition.
Therefore,scans image the bone response to the tumor and
not the tumor itself (Fig. 6-8). Even a 5% bone turnover can
be detected by bone scan. Radiographs, on the other hand,
Breast Carcinoma
Despite increased screening with mammography, a large
number of patients with breast cancer present with
advanced disease. Mean survival is only 24 months
among those with confirmed bone disease. Autopsy
studies have shown osseous metastases in 50–80% of
patients with breast carcinoma. Like prostate cancer,
stage of disease correlates with the incidence of osseous
metastases on bone scan: 0.5% in stage I, 2–3% in stage II,
8% in stage III, and 13% in stage IV. Bone scans are not
generally performed in patients with stage I or II disease.
 Skeletal Scintigraphy 121

Figure 6-8 Progression of skeletal metastases. A, Initial skeletal scintigram in a patient with
multiple skeletal metastases, including the skull. Note the intense uptake in the calvarial lesion with
a small area of decreased uptake centrally. B, Several months later, the metastatic disease has
progressed in the axial skeleton, the calvarium, and ribs. The overall diameter of the skull lesion has
increased and the central photon-deficient area is much larger. The increased uptake is in bone at
the margin of the metastatic lesion.

Although skeletal scintigraphy has a high sensitivity for

breast carcinoma, it may not detect all lesions, such as Box 6-5 Scintigraphic Patterns in
those contained in the marrow or more lytic lesions. Metastatic Disease

Lung Carcinoma Solitary focal lesions

Although up to 50% of patients dying from a primary lung
cancer have osseous metastasis at autopsy,there is no com-
plete agreement on when to use skeletal scintigraphy.
Staging is generally done with CT, surgery (including medi-
astinoscopy and video-assisted thoracoscopic surgery
[VATS]), and increasingly with F-18 FDG PET. Skeletal
scintigraphy is useful in a patient who develops pain dur-
ing or after treatment. However, it appears less useful in
cases of local and mediastinal invasion or with advanced
disease where therapy will be palliative, although it may
also be helpful in planning radiation therapy.
Scintigraphic Patterns in Metastatic Disease
The scintigraphic patterns encountered in skeletal
metastatic disease are summarized in Box 6-5. A deci-
Multiple focal lesions
Diffuse involvement (“superscan”)
Photon-deficient lesions (cold lesions)
Normal (false negative)
Flare phenomenon (follow-up studies)
Soft tissue lesions (tracer uptake in tumor)
sion tree algorithm for the workup of patients with
proven non osseous tumors is described in Fig. 6-9. The
classic pattern of metastatic disease is that of multiple,
focal lesions distributed randomly in the skeleton
(Fig. 6-10). Although this typical pattern provides a high
degree of clinical certainty as to the diagnosis, several
other etiologies can also have multiple areas of uptake
(Box 6-6). These must be differentiated from osseous
122 NUCLEAR MEDICINE: THE REQUISITES
Positive: multiple
"characteristic lesions"
Solitary lesion
High probability (or small number)
of metastatic
disease
Negative or
equivocal
Computed
tomography
or MRI
Biopsy
Figure 6-9 Simplified algorithm for the workup of patients with suspected skeletal metastasis.
metastasis. The key is to recognize the different features
and patterns of these other etiologies. Final diagnosis
may depend on correlation with anatomical imaging.
Osteoarthritic changes are routinely seen and often can
be identified by classic locations. These include the medial
compartment of the knee,hand,and wrist (especially at the
base of the first metacarpal),shoulder,and bones of the feet
(Fig. 6-11). The patella frequently shows increased uptake
due to chondromalacia and degenerative change. Arthritic
changes are frequently bilateral and on both sides of the
joint.Degenerative changes in the spine are more problem-
atic because both metastasis and arthritic changes occur in
the same location.
Caution must be exercised when assessing uptake in the
spine. The spatial resolution of planar images is not suffi-
cient to determine the region of the vertebral body
involved. SPECT may localize a lesion to the pedicle that is
the typical location of metastasis. A bone scan lesion in the
central vertebral body, even when near the end plate and
disc space, could be degenerative or malignant and may
require short term follow-up. Radiographic and CT correla-
tion often identify the etiology of the uptake as degenera-
tive changes: facet hypertrophy, disc space narrowing, and
osteophyte formation.Sometimes MRI can add useful infor-
mation in this difficult situation.
The findings of trauma can mimic the appearance of
metastasis. Patients should be closely questioned for any
history of trauma. In the ribs, a vertical alignment of focal
Whole body Negative
skeletal
scintigram
Multiple "atypical"
lesions No metastatic
disease detected:
imaging survey
complete
Standard Benign condition or
radiographs
metastatic disease
demonstrated
Diagnosis
established
abnormal uptake in several or successive ribs is classic for
trauma. The nonrandom pattern is not expected in
metastatic disease (Figs.6-12 and 6-13). A metastatic lesion
tracks along the bone as seen in Fig.6-5 rather than remain-
ing focal. Radiographic correlation may show the cortical
disruption or callous formation. Because bone scan fre-
quently detects fractures not seen on radiographs, correla-
tion with CT or short-term follow-up bone scan may be
needed if no fracture is seen on the radiograph.Persistently
positive skeletal activity from old trauma poses another
interpretive problem.
A number of other etiologies can cause multifocal
abnormalities. Infarctions in sickle cell anemia can cause
multiple areas of increased and decreased uptake.
Cushing’s disease and osteomalacia, for example, fre-
quently cause disproportionate rib lesions as compared
with other areas. Osteoporosis may result in dorsal kypho-
sis and classic fractures such as the vertebral insufficiency
fractures and the H-type fracture of the sacrum. Paget’s dis-
ease, which is addressed later in the chapter, may be differ-
entiated from metastasis by an expansion of the bone and
classic locations.
Solitary Lesions
Metastatic disease may present as a solitary bone lesion
(Fig. 6-14). The chance that a solitary lesion is due to
malignancy varies by location (Box 6-7). Uptake in a rib
in a patient with known malignancy has a 10–20% chance

Figure 6-10 Widespread metastases. Anterior (A) and
posterior (B) whole body scintigrams in a patient with widely
distributed metastatic disease. Lesions are present in the skull,
spine, ribs, pelvis, and extremities.
Box 6-6 Differential Diagnosis of
Multiple Focal Lesions (Listed in
Order of Decreasing Likelihood)
Metastatic disease
Arthritis
Trauma, osteoporotic insufficiency fractures
Paget’s disease
Other metabolic bone disease
Osteomyelitis
Numerous other conditions (fibrous dysplasia, multiple
enchondromas, infarction)
of being malignant, whereas uptake in the central skele-
ton has a much higher likelihood of being malignant.
Focal rib uptake is likely due to fracture, whereas uptake
extending along the rib is likely tumor. Common benign
causes for a solitary focus of uptake are arthritis and
trauma. Some benign bone lesions such as enchondroma,
Skeletal Scintigraphy 123
Figure 6-11 Characteristic appearance of osteoarthritis in the
hands and wrists. Uptake is increased in multiple distal
interphalangeal joints and is particularly intense at the base of the
first left metacarpal, a characteristic place for osteoarthritis.
osteoma, fibrous dysplasia, osteomyelitis, and monostotic
Paget’s disease can also cause solitary abnormalities.
Superscan
A problematic scintigraphic pattern is the “superscan” or
“beautiful bone scan.” In some patients with prostate
cancer and breast cancer, the entire axial skeleton is
involved. Uptake may be uniform enough to appear
deceptively normal (Fig. 6-15). The differential diagnosis
of the superscan pattern is provided in Box 6-8. This is
a less common interpretive problem than in the past due
to improved technology and image quality, and it should
be possible to discriminate diffuse metastasis. Classically,
absent or faint visualization of the kidneys is seen with
a superscan. However, the soft tissues (including the kid-
neys) may clear normally if the patient is imaged later
than the usual imaging time.Reviewing the available radio-
graphs on each patient will help prevent any mistake.
Flare Phenomenon
Another potentially perplexing pattern is seen in some
bone scans done on patients undergoing cyclical
chemotherapy. When a patient has a good response to
chemotherapy, the bone scan may paradoxically worsen,
with a “flare” of increased activity (Fig. 6-16). To add to
the confusion, these patients may experience increased
pain. If these lesions are followed radiographically,
increased sclerosis is seen over 2–6 months because this
is an osteoblastic response as the bone begins to heal.
This is the same time frame that the bone scan typically
shows increased uptake. The flare phenomenon rein-
forces the fact that tracer uptake is not in the tumor but
rather in the surrounding bone.
124 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-12 Typical appearance of rib fractures. A, Posterior views of the chest reveal focal
uptake in a vertical alignment in the right lower ribs and a recent left nephrectomy with resection
of some lower left ribs. B, A follow-up study 18 months later shows resolution of the right rib
uptake as the fractures healed.

Cold Lesions Extraskeletal Uptake in Soft Tissues

Lesions that are aggressive or purely lytic, as well as areas
completely replaced by tumor, may show decreased
uptake (see Fig. 6-4). A list of possible etiologies of cold
defect is provided in Box 6-9. These “cold”or photon defi-
cient areas may be difficult to spot because of overlying or
adjacent uptake. Often, these cold areas are bordered by
a rim of increased uptake. Another cause for an area of
decreased uptake is radiation therapy changes. These
are commonly geometric in shape, conforming to the
radiation port.
A number of common soft tissue neoplasms exhibit vary-
ing degrees of skeleton-seeking tracer uptake in both the
primary tumor and soft tissue metastases. The mecha-
nism of localization is thought to be a combination of
ossification in the tumor and binding to macromolecules.
The degree of uptake is not sufficient to use the Tc-99m-
labeled bone agents as primary tumor imaging agents.
Tumors most commonly seen are carcinoma of the
breast, lung, melanoma, neuroblastoma, metastatic colon
carcinoma to liver, and some malignant pleural effusions.

Figure 6-13 Multiple rib fractures bilaterally. The pattern of
linearly aligned lesions is highly characteristic for trauma and
would be unusual as a pattern for metastatic disease.
Soft tissue uptake must be differentiated from bone
pathology (Fig. 6-17 and Fig. 6-18).
Imaging Findings in Specific Tumors
Breast Carcinoma
Skeletal scintigraphy is highly sensitive in breast cancer.
Patients may show local invasion of the ribs or dissemi-
nated disease. Another important pattern to recognize is
involvement of the sternum via the internal mammary
Skeletal Scintigraphy 125
Figure 6-14 Suspected metastatic disease on bone scan. A, Pos-
terior spot views show that a solitary area of abnormally increased
uptake can be seen in the lower cervical spine. B, Computed tomo-
graphic scan obtained at the level of scintigraphic abnormality reveals
extensive destruction of the corresponding vertebral body and
demonstrates the clinically palpable soft tissue mass in the neck.
nodes. Although activity in the sternum is most often
benign, there is a high incidence of metastatic disease in
breast cancer patients (>80%). Metastatic disease to the
soft tissues may be seen (see Fig. 6-18) and increased
uptake may be present in malignant pleural effusions.
126 NUCLEAR MEDICINE: THE REQUISITES
Box 6-7 Metastatic Disease Presenting as
a Solitary Focus in Patients with
Known Cancer
Spine and pelvis 60–70%
Skull 40–50%
Rib 10–20%
Sternum (in breast carcinoma) 75%
Figure 6-15 “Superscan”of a patient with prostatic carcinoma.
In this case, the uptake is nonuniform enough that the abnormality
is easily seen. The images show an increased skeletal uptake,
largely in the axial skeleton. Soft tissue activity is decreased. The
kidneys are faint. The bladder is well-visualized, indicating that
failure to see the kidneys is not due to absence of tracer excretion
through them. Rather, the uptake in the skeleton is so intense that
the kidney activity is below the windowing threshold.
Lung Carcinoma
Interesting patterns of disease may occur on scintigrams
in lung cancer. Because these tumors can easily invade
the vasculature, arterial metastases are more common.
These tumor emboli can reach the distal extremities.
Thus, appendicular involvement is more common with
aggressive lung cancer than cancer of the breast or
prostate. Also, a characteristic periosteal change is seen
Box 6-8 Differential Diagnosis for
Superscan Pattern
COMMON
Metastases (prostate, breast)
Renal osteodystrophy
Delayed Images
LESS COMMON
Severe hyperparathyroidism (rare primary)
Osteomalacia
Paget’s disease
in lung cancer due to hypertrophic osteoarthropathy
(Figs. 6-19 and 6-20). This is most often seen as linear,
parallel “track”uptake in the medial and lateral margins of
the long bones, but may be patchy or show skip areas.
The patella, scapula, skull, clavicles, and hands and feet
may be involved.
Neuroblastoma
Neuroblastoma has a neural crest origin and is the most
common solid tumor to metastasize to bone in children
(Fig. 6-21). Tc-99m MDP scintigraphy is twice as sensi-
tive as radiographs on a lesion-by-lesion basis. MRI better
determines the extent of a lesion than bone scan. I-131 or
I-123 MIBG scanning is more sensitive than bone scan for
metastases, although the combination of both gives the
highest sensitivity.
Lesions are typically multifocal and in the metaphyses.
However, involvement in the skull, vertebrae, ribs, and
pelvis is also common. Early involvement may be sym-
metric and therefore difficult to diagnose on the bone
scan due to the normal intense activity in the ends of
growing bones.
A unique characteristic of neuroblastoma is the avid-
ity of Tc-99m diphosphonates for the primary tumor.
Approximately 30–50% of primary tumors are demon-
strated scintigraphically. Occasionally, neuroblastomas are
discovered in children undergoing radionuclide imaging
to evaluate another condition. Particular attention should
be paid to the abdomen.
Other Tumors of Epithelial Origin
Numerous other tumors metastasize to bone. The sensi-
tivity for renal cell carcinoma is low and best assessed on
skeletal survey or by MRI.Likewise,thyroid cancer is rarely
detected on skeletal scintigraphy and is better evaluated
with iodine-131 or,if noniodine avid,by F-18 FDG PET.
Gastrointestinal tract and gynecological cancers do
not commonly metastasize to bone early in their
courses. Late disease involves bone by direct exten-
 Skeletal Scintigraphy 127

Figure 6-16 Flare phenomenon. Ten-month sequence of posterior whole body scintigrams of
a patient undergoing chemotherapy for carcinoma of the breast. Note the increased intensity of
uptake in the skull, spine, and pelvis, especially between the second and third images in the
sequence. Although the scintigram appears worse, the patient was improving clinically with
reduced bone pain and radiographic evidence of healing.

sion. The success of therapy, including chemotherapy,
in controlling gastrointestinal tumors has led to an
increase in cases with skeletal involvement. Due to
longer survival and control of local and regional metas-
tases that usually cause death, bone metastases can
manifest.
Primary Tumors
Malignant Tumors
Primary bone tumors have avid uptake of the bone-seeking
radiopharmaceuticals (Fig. 6-22). These tumors include
the primary neoplasms, osteosarcoma, Ewing’s sarcoma,
and chondrosarcoma. However,skeletal scintigraphy is sel-
dom used in the workup of primary bone neoplasms
because it does not address the questions the orthopedic
surgeon needs answered: assessment of soft tissue exten-
sion or definition of tumor margins. MRI is the primary
modality for osteosarcoma evaluation (Fig.6-23). Although
most primary bone tumors are monostotic, the occasional
polyostotic involvement would be missed without a whole
body survey of some kind (Fig. 6-24). Both thallium-201
and technetium-99m sestamibi have been used for sarcoma
imaging. They may help to determine if the primary tumor
is high- or low-grade and serve as a baseline so that it can
be used to evaluate response to therapy. The increased
uptake in FDG PET of high-grade tumors has been investi-
gated but has not found a significant clinical role.
Multiple Myeloma
The most common primary bone tumor in adults is mul-
tiple myeloma. It is a tumor of the marrow and typically
involves the vertebrae,pelvis,ribs,and skull. Radiographs
128 NUCLEAR MEDICINE: THE REQUISITES

infiltration. In blast crisis, diffusely increased uptake that

Box 6-9 Differential Diagnosis of a Cold is greater at the ends of the long bones may be present.
Defect Lymphoma
Hodgkin’s disease will involve the skeleton approxi-
Metal artifact (pacemaker, prosthesis) mately one third of the time and the bone scan may show
Radiation changes focal or diffuse uptake of Tc-99m MDP. Skeletal scintigra-
Barium in bowel phy is less useful in non-Hodgkin’s lymphoma. In gen-
Vascular eral, lymphoma is best evaluated with gallium-67, F-18
Early avascular necrosis FDG PET, and CT.
Early infarct Histiocytosis
Multiple myeloma
The sensitivity of bone scan varies with the spectrum of
Osseous metastasis
Renal cell carcinoma
disease in histiocytosis. Although uptake is reliably
Thyroid carcinoma seen in eosinophilic granuloma, detection of histiocyto-
Anaplastic tumors sis is limited, with lesions seen from one-third to two
Neuroblastoma thirds of the time. Frequently, decreased uptake is
Breast and lung carcinoma shown.
Tumor marrow involvement
Lymphoma Benign Bone Tumors
Leukemia Usually, benign bone tumors are characterized by their
Benign tumors, cysts radiographic appearance. The role of skeletal scintigra-
phy is very limited in general, although understanding is
critical as benign tumors may be encountered during
imaging for other reasons. Some benign tumors have
may only show osteopenia or a permeative pattern that intense uptake similar to malignant tumors. These
can be confused with metastatic disease. Although bone include osteoid osteomas, giant cell tumors, and fibrous
scan will show some of the lesions as areas of decreased dysplasia. Most benign bone tumors are variable in
and sometimes increased uptake, radiographs are more appearance. Table 6-2 lists some of the benign bone
sensitive overall for detecting disease. The lower sensitivity tumors and their appearance on bone scan.
of bone scan relates to the lack of reactive bone forma- Osteoid Osteoma
tion in response to the lesions. Scintigraphy is very useful in identification of an osteoid
Leukemia osteoma. Classically, these lesions present in adolescents
Bone scan plays a very limited role in the evaluation of and young adults with severe pain at night. They com-
leukemia. Patients with leukemia imaged with Tc-99m monly occur in the proximal femur and spine and may
MDP may show focal increased uptake in areas of marrow be difficult to detect with conventional radiography,

Figure 6-17 A, Tc-99m MDP uptake in a right parietal stroke. B, CT confirms the etiology of the
uptake is intracranial and not in the skull.
 Skeletal Scintigraphy 129

especially in the spine. The diagnosis on radiographs can

be made if a central lucent nidus is seen surrounded by
sclerosis (Fig. 6-25). Skeletal scintigraphy is very sensitive
as the lesion will show increased uptake (Fig.6-26).SPECT
adds to this sensitivity and is particularly useful in the
spine.Surgeons can use an intraoperative probe to localize
the lesion with its increased activity. However, CT has
largely eliminated the need for the bone scan in the work
up of osteoid osteoma.
Other Indications
Bone scan may be useful in assessing an atypical bone
island on radiographs. If the sclerotic lesion on radio-
graph does not show increased scintigraphic activity, it
is unlikely to be malignant.
Osteochondromas are common cartilage-containing
benign tumors. They may show variable uptake that
diminishes as the skeleton matures. Rarely, osteochondro-
mas will show malignant transformation, usually into
a chondrosarcoma. This degeneration occurs less than 1%
of the time in a solitary lesion and less than 5% of the time
in hereditary multiple osteochondromatosis (hereditary
multiple exostoses). Although bone scan can exclude
malignancy if no increased uptake is seen, the presence of
increased activity does not differentiate benign from
malignant lesions. Any new uptake in a lesion that previ-
ously had none is suspicious. Osteochondromas are usu-
ally best evaluated with CT and MRI.
Enchondromas usually present as a cystic lesion in the
hands or a sclerotic area reminiscent of a bone infarct
elsewhere. They are benign but can degenerate into
a malignant tumor. This degeneration is more common
in multiple enchondromatosis (Ollier’s disease). Bone
scan may help identify multiple lesions, but the role is
otherwise very limited.
Bone Dysplasias
Numerous bone dysplasias demonstrate increased skele-
tal tracer uptake (Figs. 6-27 and 6-28). Fibrous dysplasia
is the most commonly encountered of these and may be
monostotic or polyostotic (see Fig. 6-28). The degree of
increased tracer uptake is typically high, rivaling that
seen in Paget’s disease. Distinguishing features are the
younger age of the patient and the different pattern of
involvement.When Paget’s disease involves a long bone,
it invariably extends to at least one end of the bone. Figure 6-18 Soft tissue metastasis in breast cancer. Posterior
Fibrous dysplasia frequently does not involve the end of scintigram of a patient with breast carcinoma. The patient has
the bone. Other dysplasias associated with increased multiple skeletal metastases. Uptake is intense in the soft tissue liver
tracer uptake are listed in Box 6-10. metastasis. The uptake projects just superolateral to the right kidney.
130 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-19 Hypertrophic osteoarthropathy in a patient with bronchogenic lung carcinoma.

A, Planar whole body scintigrams reveal the classic pattern of uptake in the periosteal region of the
long bones. B, Follow-up scan 9 months later shows increased activity in a treated left apical lung
mass. Radiation therapy changes of decreased uptake in the upper thoracic spine are seen. With
successful treatment, the findings of hypertrophic osteoarthropathy have resolved. C, Spot views of
the femurs more clearly show the abnormal uptake (left) that later resolves more clearly (right).
 Skeletal Scintigraphy 131

Figure 6-20 A, Florid hypertrophic osteoarthropathy. The bones of the upper and lower
extremities are diffusely involved, as are the clavicles, mandible, and skull. Although the pattern may
be confusing, the patient did not have skeletal metastatic disease. Involvement of the extremities is
one clue. B, Chest x-ray reveals a bronchogenic carcinoma in the right upper lobe just above the
right hilum.
Continued

costochondral junctions. Striking “beading” of the cos-

Metabolic Bone Disease
A number of metabolic conditions can result in marked
bone scan abnormalities ( Box 6-11). The patterns of
metabolic bone disease must be recognized to avoid
confusion with pathology such as metastatic disease.
However, there is no role for bone scan in the diagnosis
and management of these diseases.
Inadequate osseous mineralization results in osteo-
malacia. This gives the bones a washed-out, chalky
appearance with decreased trabeculae on radiographs.
Osteomalacia, hyperparathyroidism, renal osteodystro-
phy and hypervitaminosis D can all cause patterns of
increased scintigraphic uptake.
Problems with vitamin D metabolism can cause rick-
ets. Although this affects adults, the changes are most
striking in a growing skeleton. Radiographs may show
changes at the growth plates ( frayed, widened, cup-
ping) of the proximal humerus, distal femur, distal tibia,
and distal forearm. Significant changes are seen in the
tochondral junction may be seen (Fig. 6-29).
Hyperparathyroidism
Hyperparathyroidism, renal osteodystrophy, osteomala-
cia, and hypervitaminosis D can all result in various
patterns of increased uptake in the skeleton, and
Generalized increased tracer uptake may be seen
throughout the skeleton ( Fig. 6-30; Box 6-12). An
increased skeleton-to-soft tissue ratio and poor renal
visualization may create an appearance similar to
a superscan seen in metastatic disease.
The bone scan in long standing renal osteodystro-
phy often has the most extreme appearance. Increased
activity is typically seen in the periarticular regions,
skull, mandible, maxilla, and sternum. These same
findings may be seen to a varying degree in secondary
hyperparathyroidism and osteomalacia. For example,
pseudofractures are often present in osteomalacia
and demonstrate avid radiopharmaceutical uptake
132 NUCLEAR MEDICINE: THE REQUISITES
Figure 6-20, cont’d C, Radiograph of the femurs shows
characteristic periosteal new bone bilaterally on both the medial
and lateral aspects of the femoral shaft.
(Fig. 6-31). Extraskeletal uptake may be seen. Soft tissue
uptake in the lungs may be caused by an increased
serum calcium-phosphate ratio. In hyperparathyroidism
associated with renal failure, a classic pattern of uptake
is seen in the lungs, stomach, and kidneys. These organs
are all involved in acid-base metabolism (Fig. 6-32).
Osteoporosis
Although inadequate osseous mineralization results in
osteomalacia, osteoporosis is a decrease in bone mass.
This results in an increase in fragility due to decreased
bone mineral content and architectural deterioration.
On radiographs, the bones have a washed-out appear-
ance with decreased trabeculae.
Skeletal scintigraphy does not have a role in the diag-
nosis of osteoporosis but is useful in surveying the
entire skeleton for osteoporotic insufficiency fractures.
Because these may be asymptomatic, the ability to sur-
vey the entire skeleton is advantageous. Compression
fractures of the spine are common and may show abnor-
Figure 6-21 Bone metastases in neuroblastoma. Abnormal
tracer localization is present in both femurs and distal tibial
metaphyses, more extensive on the left than the right.
mal uptake before radiographic changes (Fig. 6-33).
These insufficiency osteoporotic fractures may show
persistent uptake for months or years, so it can be diffi-
cult to determine if the fracture is acute.
Sacral insufficiency fractures are also common and are
often difficult to diagnose radiographically or by CT. The
most common pattern is the H or butterfly pattern, with
a horizontal band of increased uptake across the body of
the sacrum and two vertical limbs of activity in the sacral
alae (Fig. 6-34). Several pattern variations may be seen,
including asymmetry of the alar activity. Less severe frac-
tures may show only horizontal linear uptake.
Paget’s Disease
Paget’s disease may be included in the metabolic bone
disease category, although the exact etiology is not
entirely understood. A chronic bone disease of the eld-
erly, it causes enlarged, coarsened bones. The diagnosis
can usually be made by radiographs, although CT and
MRI can be used to assess complications of Paget’s.
 Skeletal Scintigraphy 133

Figure 6-22 Osteosarcoma of the right distal femur. The

degree of tracer accumulation in the lesion is striking. The
increased blood flow induced by the osteosarcoma results in
increased tracer delivery to the entire limb. This “extended”or
augmented pattern of uptake adds to the difficulty in using the
skeletal scintigram to determine the margins of primary bone
tumors. (The focal activity over the right ribs is a marker.)

A bone scan is useful for the evaluation of the extent of

disease. When it is found incidentally, the patterns of
Paget’s disease must be recognized.
The scintigraphic appearance is striking,with intensely
increased tracer localization (Figs. 6-35 and 6-36). The
expansion of bone demonstrated radiographically is not
well-assessed by scintigraphy, owing to the lower resolu-
tion of the technique and the “blooming” appearance of
very intense areas of uptake, but it is certainly suggested
on the images. The pelvis is the most commonly involved
site, followed by the spine, skull, femur, scapula, tibia, and
humerus. The increased uptake is seen in all phases of
the disease: the early resorptive, mixed, and sclerotic
phases.In osteoporosis circumscripta,a characteristic rim
of increased uptake borders the lesion. Figure 6-23 A, Osteosarcoma of the left distal femur and
a metastatic lesion in the left proximal femur. B, Coronal T1-
weighted MRI. Superior anatomical information about the osseous
Skeletal Trauma and soft tissue extent of the tumor. However,it missed the second
lesion in the proximal femur which was out of the field of view.
Skeletal scintigraphy is able to demonstrate abnormali-
ties early after direct trauma (Figs. 6-12 and 6-37). The Detection of Fractures
bone scan findings from trauma may persist for quite Approximately 80% of fractures can be visualized by 24
some time. This may create problems when using skele- hours after trauma. The earliest scintigraphic appear-
tal scintigraphy for another purpose, such as the detec- ance is diffusely increased uptake, most likely the result
tion of metastatic disease. of hyperemia at the fracture site. After 3 days, 95% of
134 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-24 A–B, Extraosseous osteosarcoma arising in the right medial thigh area. The tumor
is widely disseminated with skeletal metastases, soft tissue metastases, and pulmonary metastases.
This study is a dramatic example of the ability of skeletal scintigraphy to survey the entire body.

fractures are positive on scintigraphy; in patients under
the age of 65, essentially all fractures are positive by this
time. Advanced age and debilitation are factors con-
tributing to a lack of visualization or delayed visualiza-
tion of fractures. The maximum degree of fracture
uptake occurs 7 or more days after trauma and delayed
imaging in this time frame is recommended in difficult
or equivocal cases. In the past, skeletal scintigraphy was
frequently performed to evaluate radiographically
occult hip fractures. This role has largely moved to MRI,
which is highly sensitive and often provides additional
information.
The time a fracture takes to return to normal on the
bone scan depends on its location, its stability, and the
 Skeletal Scintigraphy 135

Table 6-2 Benign Bone Lesions on Skeletal Scintigraphy

Degree of Uptake Malignancy Potential Comments

INTENSE UPTAKE
Aneurysmal bone cyst No Donut sign pattern
Chondroblastoma Almost always benign Bone scan positive does not diagnose
malignancy
Giant cell tumor 10%
Fibrous dysplasia <1%
Osteoma No Gardner’s syndrome
Osteoid osteoma No Osteoblastoma >2 cm

ISOINTENSE/MILD UPTAKE
Bone island No
Enchondroma Solitary: <20% long bones
Multiple enchondromatosis: <50%
Nonossifying fibroma No

VARIABLE UPTAKE
Osteochondroma <1%

HEREDITARY MULTIPLE EXOSTOSES

Hemangioma No Prominent trabeculae diagnostic
Eosinophilic granuloma No; least aggressive histiocytosis Monostotic /Polyostotic
group

LOW UPTAKE
Unicameral bone cyst No

degree of damage to the skeleton. Some 60–80% of
nondisplaced uncomplicated fractures revert to nor-
mal in 1 year and over 95% revert in 3 years (Table 6-3).
However, there are many instances in which dis-
placed fractures remained positive indefinitely (e.g.,
involvement of a joint by posttraumatic arthritis causes
prolonged abnormal uptake). Patients undergoing
metastatic skeletal survey should be routinely asked
about prior trauma. In a prospective study by Kim, nearly
half of patients being evaluated for skeletal metastatic
disease reported previous fractures.In all,26% of the frac-
ture sites were positive at the time of scintigraphic exam-
ination, including 16 (16%) of 98 sites where the trauma
had occurred more than 5 years before. Structural defor-
mity and posttraumatic arthritis were the most common
reasons for prolonged positive studies.
Iatrogenic Trauma
Iatrogenic trauma to either the skeleton or soft tissues may
result in abnormal uptake on the bone scan. The key to
correct interpretation is an accurate history. Craniotomy
typically leaves a rim pattern at the surgical margin that
may persist for months postoperatively. Rib retraction dur-
ing thoracotomy can elicit periosteal reaction and
increased uptake without actual resection of bone being
involved. Bone resections are recognized as photon-defi-
cient areas, although small laminectomies are usually not
appreciated scintigraphically.
Areas of the skeleton receiving therapeutic levels of
external beam ionizing radiation (typically ≥4000 rads)
characteristically demonstrate decreased uptake within
6 months to 1 year after therapy. The threshold for the
effect is on the order of 2000 rads. The mechanism is
probably decreased osteogenesis and decreased blood
flow to postirradiated bone. The scintigraphic hallmark
is a geometrical pattern of regionally decreased tracer
uptake (see Fig. 6-19B). An increase in uptake immedi-
ately after therapy may be seen due to hyperemia.
Child Abuse
The generally high sensitivity of skeletal scintigraphy
would seem to make it an ideal survey test in cases of sus-
pected child abuse. In practice,however,the sensitivity is
somewhat disappointing. This probably relates to the
timing of injury in relation to scintigraphy. Older frac-
tures may have healed and may not be seen scintigraphi-
cally, and skeletal scintigraphy has been reported to miss
calvarial fractures in young children. In child abuse,
radiographic skeletal survey is more sensitive than bone
scintigraphy because of its ability to demonstrate old frac-
tures. Scintigraphy should be reserved for cases of sus-
pected child abuse in which radiographs are unrevealing.
136 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-25 Osteoid osteoma: radiograph. A, Conventional tomogram of the right proximal
femur reveals a characteristic radiolucent nidus surrounded by sclerotic bone. B, Specimen
radiograph confirms the complete excision of the nidus.

Figure 6-26 Osteoid osteoma: pinhole images. Internal and external rotation pinhole spot views
of the proximal femur in a patient with suspected osteoid osteoma. An area of abnormally increased
uptake demonstrated just lateral to the lesser trochanter confirms the clinical suspicion.
 Skeletal Scintigraphy 137

Figure 6-27 Melorheostosis. A, Posterior whole body image showing intensely increased
uptake in a somewhat patchy distribution involving the right femur. B, Radiograph of the distal
femur reveals the characteristic intensely sclerotic lesion of melorheostosis, often characterized as
having the appearance of dripping candle wax.

In these cases, symmetric positioning of the patient
is critical.
Complex Regional Pain Syndrome
Complex regional pain syndrome, previously known as
reflex sympathetic dystrophy, is a complex disorder with
a variable presentation. It is an exaggerated response to
injury and immobilization with sensory, motor, and auto-
nomic features. Typically, patients present with pain,
edema, and muscle wasting in an affected extremity.
Scintigraphic findings are variable depending on the
stage of disease. Early disease (up to 5–6 months) usually
shows increased blood flow. Later in the course of disease,
blood flow may be normal or decreased. The classic pat-
tern has been described as a unilateral increase in flow and
blood pool activity with increased periarticular uptake on
delayed images. This pattern is seen less than 50% of the
time but provides the highest diagnostic accuracy. The
periarticular uptake on delayed images is found in most
cases (>95%), although specificity is lower if this finding is
138 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-28 Fibrous dysplasia. A, Uptake is markedly increased in the distal humerus, most
of the forearm, and focal areas in the hand. B, Corresponding radiograph of the left elbow reveals
characteristic expansile lesions of fibrous dysplasia.

Stress Fractures
BOX 6-10 Bone Dysplasias Associated
A significant change in activity level or a repetitive activity
with Increased Skeletal Tracer
may lead to injury to the bone. The reaction to this injury
Uptake
is remodeling of the bone. Cortical bone may become
weakened and buttressed by periosteal and endosteal
Fibrous dysplasia new bone. The final result of imbalance between resorp-
Osteogenesis imperfecta tion and replacement is a stress fracture (Table 6-4). If
Osteopetrosis
the process causing injury is allowed to continue to the
Progressive diaphyseal dysplasia (Engelmann’s disease)
point of overt fracture, healing predictably takes several
Hereditary multiple diaphyseal sclerosis (Ribbing’s
disease) months or more, compared with the several weeks
Melorheostosis required for healing of an early stress reaction.
Therefore, prompt diagnosis and appropriate change in
activity is critical.
seen without the increased blood flow.Infection and arthri- Skeletal scintigraphy is exquisitely sensitive to the
tis could cause false positives. Some variants have been remodeling process and typically shows abnormalities at
found, including cold or decreased uptake in some adults. least 1–2 weeks before the appearance of radiographic
Children often have normal or decreased uptake. changes in stress fractures. The characteristic scinti-
 Skeletal Scintigraphy 139

Box 6-11 Metabolic Bone Disorders

OSTEOPOROSIS
Primary (Idiopathic)
Senile, postmenopausal

Secondary
Disuse
Drugs Corticosteroids, chemotherapy, anticonvulsants
Endocrine Hyperthyroidism, primary hyperparathyroidism, Cushing’s disease, hypogonadism

OSTEOMALACIA
Vitamin D Vitamin D deficiency, hereditary disorders of vitamin D metabolism
Decreased calcium Calcium malabsorption, inadequate intake, calcitonin secreting tumors
Phosphate loss Renal tubular disease, hemodialysis, transplant
Other Liver disease, phenytoin, prematurity

HYPERPARATHYROIDISM
Primary Parathyroid adenoma, parathyroid hyperplasia
Secondary Chronic renal insufficiency, phosphate metabolism abnormalities, parathyroid
hyperplasia
Tertiary Autonomous parathyroid glands from long standing secondary hyperparathyroidism

RENAL OSTEODYSTROPHY Chronic renal failure

HYPOPARATHYROIDISM Iatrogenic loss/damage parathyroid glands during thyroidectomy;
pseudohypoparathyroidism genetic end-organ resistance
METAL TOXICITIES
Aluminum-induced bone disease
Fluorosis
Heavy metal poisoning

graphic appearance is that of intense uptake at the frac-
ture site. The configuration is oval or fusiform with the
long axis of increased uptake parallel to the axis of the
bone (Fig. 6-38).
MRI is currently the modality most commonly used to
evaluate stress fracture. It can provide additional infor-
mation such as the status of soft tissues and tendons.
MRI does not expose patients to ionizing radiation. MRI
can identify marrow edema early in the stress response,
but edema alone is nonspecific. MRI may be helpful by
identifying a true fracture in an area of edema. Fractures
are visualized as linear abnormalities on T1 scans.
Spondylolysis occurs in the lumbar spine in the pars
articularus. This is often seen due to repetitive trauma
in young athletes. Most commonly the abnormality
occurs at L4–5. In some instances, all examinations
may be normal, including MRI and radiographs, but a
SPECT study may reveal the pathology (Fig. 6-39).
Shin Splints
The term shin splints is applied generically to
describe stress-related leg soreness. Patients complain
of mild to moderate exercise-induced pain along the
medial or posteromedial aspect of the tibia. In nuclear
medicine, the term is now used to describe a specific
combination of clinical and scintigraphic findings.
Increased tracer uptake is seen on the scintigram, typi-
cally involving a large portion of the middle to distal
tibia ( Fig. 6-40). Most cases are bilateral although not
necessarily symmetrical. The radionuclide uptake is
superficial, only mild to moderate in intensity, and
lacks a focal aspect seen with true stress fractures.
Hyperemia is limited, unlike stress fractures which
show intense hyperemia.
A phenomenon perhaps related to shin splints is
activity-induced enthesopathy. The term simply refers
to a disease process at the site of tendon or ligament
attachment to bone. In athletes, repeated microtears
with subsequent healing reaction can result in increased
tracer uptake at corresponding locations. Osteitis pubis,
plantar fasciitis, Achilles tendonitis, and some cases of
pulled hamstring muscles are examples. A periosteal
reaction develops at the site of stress, resulting in
increased skeletal tracer localization.
140 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-29 Renal osteodystrophy. Whole body (A) and spot images (B) of a patient with long-
standing renal failure show classic skeletal changes of severe renal osteodystrophy. Abnormal
activity is seen in the face and skull as well as the distal ends of the long bones. The rib tip activity
has been called the rachitic “rosary bead”configuration. Focal uptake in the left scapula was a
fracture although brown tumors can have a similar appearance.
 Skeletal Scintigraphy 141

Figure 6-30 Renal osteodystrophy. A–B,The absence of soft tissue uptake is striking with an
appearance similar to the “superscan”seen in metastatic disease. The prominent rib end activity may
help differentiate the two etiologies. The native kidneys had failed,and a renal transplant is noted in the
right iliac fossa.
Continued
142 NUCLEAR MEDICINE: THE REQUISITES

tissue provides a site for radionuclide deposition when

combined with phosphate.
The scintigraphic pattern reflects the muscle groups
undergoing injury.In marathon runners,the most striking
uptake is usually in the muscles of the thigh.
Rhabdomyolysis induced by renal failure is generally dif-
fuse. The time course of scintigraphic abnormality
appears to be similar to that for acute myocardial infarc-
tion. The greatest degree of uptake is seen at 24–48
hours following injury. The changes resolve by 1 week.

Heterotopic Bone Formation

Heterotopic bone formation can occur in the muscle due
to numerous conditions. It is most often a direct result of
trauma to the muscle in myositis ossificans (Fig. 6-42).
However,it can also be a serious problem in paralyzed mus-
cles and prolonged immobilization. The bone scan in these
patients will reveal increased Tc-99m MDP deposition in the
muscles on delayed images. Often, the blood flow and
immediate blood pool images show more intense activity
than the delayed images. Increased soft tissue uptake on
a bone scan typically occurs long before any radiographic
change. If patients are treated in these early stages,they may
avoid more severe and lasting complications such as severely
contracted and ossified muscles at the hips in paraplegia.

BONE INFARCTION AND

Figure 6-30, cont’d C, Increased activity in the skull and
sternum may be especially prominent. Note the increased axial
skeletal uptake and paucity of soft tissue background activity.
Rhabdomyolysis
Another athletic injury that is seen in this day of
marathons and triathlons is rhabdomyolysis (Fig. 6-41).
The localization of skeletal tracers in exercise-damaged
skeletal muscle is probably similar to the localization in
damaged myocardium. Calcium buildup in damaged
OSTEONECROSIS
Necrosis of the bone has numerous causes (Box 6-13).
Because this is an evolving process,the appearance on skele-
tal scintigraphy depends greatly on the time frame in which
imaging is performed. With acute interruption of the blood
supply, newly infarcted bone appears scintigraphically cold
or photon-deficient. In the postinfarction or healing phase,
osteogenesis and tracer uptake at the margin of the infarcted
area are increased. Skeletal scintigrams can show intensely
increased tracer uptake during the healing period.

Legg-Calvé-Perthes Disease
Box 6-12 Distribution of Increased Legg-Calvé-Perthes disease most commonly affects chil-
Skeletal Uptake in dren between the ages of 5–9 years with predominance
Hyperparathyroidism in boys (4:1 to 5:1). It is a form of osteochondrosis and
results in avascular necrosis of the capital femoral epi-
physis. The mechanism of injury is unknown except
Diffuse axial that the vascular supply of the femoral head is thought
Periarticular
to be especially vulnerable in the most commonly
Skull
affected age group.
Mandible, fascial bones
Costochondral junctions The best scintigraphic technique for detecting the
Sternum abnormality in the femoral head is to use some form of
Lungs magnification and to image in the frogleg lateral projection.
Stomach Classically, early in the course of the disease before healing
has occurred, a discrete photon-deficient area can be seen
Figure 6-31 Osteomalacia. Anterior (A) and posterior (B) views. The patient was referred to
rule out metastatic disease. The unusually large number of rib lesions raised the suspicion of
metabolic bone disease rather than metastases.

Figure 6-32 Hyperparathyroidism. Whole body views show diffusely increased uptake in the
lungs and stomach.
 Figure 6-33 Osteoporosis on surveillance images obtained several months apart. A, The initial
study shows a single vertebral compression fracture caused by osteoporosis involving the lower
thoracic spine. B, The subsequent study shows healing with normalization of uptake in the initial
abnormality. Three new compression fractures are seen in the thoracic and lumbar spine.

Figure 6-35 Monostotic Paget’s disease involving the right

Figure 6-34 Sacral insufficiency fracture. Posterior spot view
of a patient with osteoporosis. The patient has an H-type pattern
with a horizontal band of increased uptake across the body of the
sacrum and bilaterally increased uptake in the sacral alae.
distal femur. The uptake is extremely intense, with the appearance
of bony expansion. The observation about expansion must be
made with caution because of the extreme intensity of uptake and
“blooming”of the recorded activity. Also note involvement of the
ischium.
 Skeletal Scintigraphy 145

Figure 6-36 Multifocal Paget’s disease. A, Abnormal uptake in the left hemipelvis, upper lumbar
spine and, to a lesser extent, right hip in a patient with Paget’s disease. When Paget’s disease
involves the axial bones, it must be differentiated from metastatic disease by the location and bone
expansion. Radiographic correlation will show the typical coarsened trabeculae. B, When the sites
are more numerous, the diagnosis is obvious based on the typical distribution of lesions.

in the upper outer portion of the capital femoral epiphysis
with a lentiform configuration (Fig. 6-43). Areas of photon
deficiency are well demonstrated by SPECT imaging.
As healing occurs, increased uptake is first seen at the
margin of the photon-deficient area, and gradually the
scintigram demonstrates filling in of activity. In severe
cases, the femoral head never reverts to normal. Increased
tracer uptake is seen for a prolonged period—many months
or more.
Currently MRI is the imaging modality of choice for
the evaluation of Legg-Calvé-Perthes disease, as well as
other causes of osteonecrosis. Compared with nuclear
scintigraphy, MRI has comparable or higher sensitivity
and higher specificity. MRI also provides a range of addi-
tional information, including evaluation of articular carti-
lage, detection of acetabular labral tears, and visualization
of metaphyseal cysts that are indicators of prognosis.
Steroid-Induced Osteonecrosis
Skeletal scintigraphy rarely shows photon deficient areas
in steroid-induced osteonecrosis. The vast majority of
146 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-37 Trauma to the distal extremity. A, Skeletal scintigram of a patient who had
sustained direct trauma to the right foot and ankle reveals multiple focal areas of abnormal tracer
accumulation from fracture. B, The radiograph illustrates fractures of the base of the fifth metatarsal
and lateral cuneiform.

cases show increased radiotracer uptake. Although the

Table 6-3 Skeletal Scintigraphy in Trauma: Time pathogenesis of steroid-induced osteonecrosis is still
Course from Fracture to Return to Normal
being debated, it is a chronic process manifested by
microfractures and repair. The net effect most often
Fracture type and site Percent of normal seen scintigraphically is increased tracer localization.
Nonmanipulated Closed
Fractures* 1 year 3 years
Sickle Cell Anemia
Vertebra 59 97 Skeletal scintigrams in patients with sickle cell anemia
Long bone 64 95
have a number of characteristic features that suggest the
Rib 79 100
diagnosis (Fig. 6-44). In the skull, the expanded marrow
All Fractures† <1 year 2–5 years >5 years space results in bilaterally increased calvarial uptake of
tracer.In the extremities,patients usually have greater rela-
All sites 30 62 84
tive uptake compared with the axial skeleton than is seen
*Adapted from Matin P: The appearance of bone scans following fractures, in normal subjects. This increased uptake may be related
including immediate and long term studies, J Nucl Med 20:1227-1231, 1979. to the persistence of hematopoietic elements throughout
†Adapted from Kim HR,Thrall JH, Keyes JW Jr: Skeletal scintigraphy following
the extremities, including the hands and feet, of patients
incidental trauma, Radiology 130:447-451, 1979.
with sickle cell anemia. As noted earlier, in normal adults
the red marrow extends only to the proximal portions of
Table 6-4 Sequence of Findings in Stress Reaction the femurs and humeri. The overall skeleton-to-back-
ground ratio is usually good and is accentuated by the
increased appendicular uptake.
Clinical
findings X-ray Scintigram In many patients with sickle cell anemia, the kidneys
appear somewhat larger than normal, which may be
Normal (resorption − − − related to a defect in the ability to concentrate urine.
= replacement)
Avid accumulation of skeletal tracer is sometimes seen in
Accelerated remodeling +/− − +
(resorption > replacement) the spleen, presumably because of prior splenic infarc-
Fatigue (resorption + +/− +++ tion and calcification (Fig. 6-45).
>> replacement) Infarctions in bone and bone marrow result in both
Exhaustion (resorption ++ + ++++ acute and long-standing changes. If the involvement
>>> replacement)
is primarily in the marrow space, the skeletal scinti-
Cortical fracture ++++ ++++ ++++
gram may not reveal the extent of the lesion as it
Adapted from Roub LW, et al: Bone stress: a radionuclide imaging perspective, does not involve the cortex where Tc-99m MDP binds.
Radiology 132:431-438, 1979. The image may be normal acutely.Within a few days as
 Skeletal Scintigraphy 147

Immediate Immediate 3-hr delay

A
Figure 6-38 Stress fracture. A, Three-phase skeletal scintigram of the feet in plantar view reveals
marked early hyperemia to the left midfoot. The immediate views (lower left and middle) reveal
increased uptake in the same area. The 3-hour delayed view shows marked focal uptake
corresponding to the base and shaft of the second metatarsal, compatible with stress fracture.
B, A radiograph performed later confirmed the fracture with callus formation (arrow) in the base
of the second metatarsal.

healing begins, the scan typically demonstrates increased
uptake.
MRI can demonstrate marrow infarctions immedi-
ately. Bone marrow scans using Tc-99m sulfur colloid are
also sensitive and are positive immediately after the
infarct (Fig. 6-46). Affected areas fail to accumulate tracer
and are seen as cold or photon deficient. The presence of
chronic marrow defects from prior bone marrow infarc-
tions persist. Thus the significance of a photon-deficient
area on marrow scanning is somewhat uncertain unless a
recent baseline study is available for comparison. Here
again, MRI has an advantage in distinguishing acute from
chronic changes.
The correlation between the marrow scan and the
bone scan is also important in the differentiation of acute
osteomyelitis from infarct. If the marrow shows a defect
in the region of increased bone scan activity, it is consis-
tent with infarct. If the marrow shows no change, then
any increased activity on the bone scan in an acute situa-
tion is most likely osteomyelitis.
OSTEOMYELITIS
Acute hematogenous osteomyelitis typically begins by
seeding of the infectious organism in the marrow space.
The untreated process extends through Volkmann canals
horizontally and in the Haversian canal system axially. The
skeletal scintigram is almost invariably abnormal by the
time clinical symptoms develop.Increased tracer uptake is
the typical finding (Fig. 6-47). Numerous studies in the lit-
erature document the superior sensitivity of skeletal
scintigraphy compared with conventional radiography in
the diagnosis of acute hematogenous osteomyelitis.
In children, Staphylococcus aureus is the most com-
mon organism and is probably responsible for 50% or
148 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-39 Spondylosis. A, Posterior scintigram from a child with low back pain after athletic
injury. Uptake is bilaterally increased at L5. B, Corresponding coronal view from a SPECT study
reveals a characteristic pattern for spondylolysis. C, Comparison radiograph reveals defect in the
pars interarticularis corresponding to the area of abnormal uptake on the scintigram.
 Skeletal Scintigraphy 149

Figure 6-40 Shin splint. Lateral views of a patient

demonstrating the classic finding of increased tracer uptake along
the posterior and medial aspects of the tibia on the left. A similar
pattern can be seen on the right, but with the addition of a focal
area distally, which could indicate stress fracture.

Figure 6-42 Myositis ossificans. Anterior and posterior whole

body scintigrams with extensive myositis ossificans involvement.

Box 6-13 Etiologies of Aseptic Bone

Necrosis

Trauma (accidental, iatrogenic)

Drug therapy (steroids)
Hypercoagulable states
Hemoglobinopathies (sickle cell disease and variants)
After radiation therapy (orthovoltage)
Caisson disease
Osteochondrosis (pediatric age group; Legg-Calvé-
Perthes disease)
Polycythemia
Leukemia
Gaucher’s disease
Alcoholism
Figure 6-41 Rhabdomyolysis. Anterior view of the thighs in a Pancreatitis
patient who recently competed in a marathon. Bilateral soft tissue Idiopathic
uptake is compatible with rhabdomyolysis.
150 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-43 Legg-Calvé-Perthes disease. A, (top row). Scintigrams by standard parallel-hole

collimator fail to reveal the abnormality Pinhole images of the same patient (bottom row) reveal the
characteristic lentiform area of decreased uptake on the left. B, Corresponding radiograph done
months later reveals deformity of the left femoral epiphysis with flattening, increased density, and
increased distance between the epiphysis and the acetabulum.

Figure 6-44 Sickle cell disease. Anterior and posterior whole
body views. Calvarial uptake is increased with relative thinning at
the midline. There is prominent skeleton to soft tissue uptake.
The kidneys appear large, and the spleen shows intense uptake.
Uptake in the knees and ankles is greater than expected for an
adult subject. The photon-deficient areas in the right femur are
due to bone and bone marrow infarction.
more of cases. The skeletal infection is commonly asso-
ciated with some other staphylococcal infection, often of
the skin. Enteric bacteria and Streptococcus are also
important pathogens.
Osteomyelitis in adults may occur by hematogenous
spread or by direct extension in an area of cellulitis. The
foot is a common site of direct extension in diabetic
patients, whereas the spine is commonly involved by the
hematogenous route. In nondiabetic adults, the axial
skeleton is more commonly involved than the appendic-
ular skeleton (Fig. 6-48).
Osteomyelitis may involve any skeletal structure.When
vertebrae are involved, the organisms may be carried
through the perispinous venous plexus, producing
involvement at multiple levels. When the vertebral end-
plates are involved, the infection may extend into the disc
Skeletal Scintigraphy 151
space causing diskitis. Diskitis is characterized by narrow-
ing of the disc space and increased uptake in the adjacent
vertebral bodies from osteomyelitis of the adjacent bone.
In some patients, especially children, increased pres-
sure in the marrow space or thrombosis of blood vessels
results in paradoxically decreased tracer uptake and
a cold or photon-deficient lesion. False-negative scinti-
graphic studies are unusual but have been reported in
infants under the age of 1 year. Other causes of false-neg-
ative examinations are imaging very early in the course
of disease and failure to recognize the significance of
photon-deficient areas.
For the diagnosis of osteomyelitis, MRI is typically per-
formed with gadolinium enhancement. Again, MRI is
very sensitive but is limited as a survey technique. If clin-
ical findings point to a specific location, MRI is useful.
When polyostotic disease is suspected or clinical find-
ings are not well-localized, skeletal scintigraphy remains
a useful survey technique. The characteristic appear-
ance of osteomyelitis by gadolinium-enhanced MRI is an
area of enhancement, whereas the centers of abscesses
do not enhance. MRI is not as useful in evaluating the
feet of diabetic patients because of changes such as
Charcot’s disease. The underlying changes and poor
blood flow are nonspecific and make diagnosis difficult.
Scintigraphic Findings of Osteomyelitis
Dynamic (or three-phase) imaging is a special technique
used in the differential diagnosis of cellulitis and
osteomyelitis (see Box 6-2). This is an important differ-
ential diagnosis for diabetic patients who have a high
incidence of both problems because of the therapeutic
implications of prolonged treatment when osteomyelitis
is diagnosed.
The key diagnostic criteria of osteomyelitis are shown
in Box 6-14. Typical osteomyelitis appears positive on all
three phases of the study. Early or arterial hyperemia
with focally, and possibly diffusely, increased radiotracer
uptake on blood pool images is seen. Progressive focal
accumulation then occurs in the involved bone on
delayed images (Fig. 6-49). Cellulitis, on the other hand,
typically demonstrates delayed or venous phase hyper-
emia and increased blood pool images activity but no
focally increased uptake in bone on delayed images
(Fig. 6-50).
Although the technique has been shown useful in dis-
tinguishing cellulitis from osteomyelitis, bone scan is not
specific. The same sequence of image findings seen in
osteomyelitis can be seen in neuropathic joint disease,
gout, fractures (including stress fractures), and rheuma-
toid arthritis, among other conditions (Box 6-15).
Improved specificity can be achieved by comparing the
three-phase bone scan to radiographs.
152 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-45 Sickle cell crisis. Posterior views with intensity set high obtained at the
time of onset of acute chest pain (A) and several days later (B) from a patient with sickle cell
anemia. The initial image reveals no abnormality in the ribs. The follow-up image
demonstrates increased uptake, particularly in the right ribs, associated with healing of the
infarctions. Note the uptake in the spleen on both images. Typically, once the spleen is
visualized scintigraphically, it remains positive.

loosened prosthesis and infection is better made with

Prosthesis Evaluation
Numerous attempts have been made to use skeletal
scintigraphy in the evaluation of patients after total joint
replacement or implantation of other metallic prosthe-
ses. These patients often experience severe pain. The
distinction between component loosening and infection
is critical in guiding management.
The findings on skeletal scintigraphy are not specific
enough for a reliable distinction between loosening of
a prosthesis and infection (Fig. 6-51). Reactive bone
around a loose prosthesis may be indistinguishable from
increased tracer uptake resulting from osteomyelitis. In
cases of a loose prosthesis, uptake is usually increased in
the region of the greater and lesser trochanters and at
the tip of the prosthesis. Loosening may also be seen as
diffusely increased uptake around the acetabular com-
ponent. This increased activity is presumably due to
remodeling of bone in response to movement of the
prosthesis. Some increased uptake is expected as a nor-
mal healing response for 1 year after placement of
a cemented prosthesis and for 2–3 years after placement
of a noncemented prosthesis. However, increased activ-
ity may persist indefinitely. Although seldom per-
formed, a baseline study 6 months to 1 year following
surgery is very useful for future comparison.
In osteomyelitis, activity is increased in the bone sur-
rounding the prosthesis. However, a negative bone scan
is useful because it helps rule out both osteomyelitis and
prosthesis loosening. The differential diagnosis between
tracers such as radiolabeled white blood cells.
Detection of an infected prosthesis with In-111 or
Tc-99m HMPAO-labeled white blood cells offers the high-
est sensitivity and specificity. This tracer localizes in
areas of infection and not in areas of remodeling or reac-
tive bone. Use of labeled white blood cells has three pit-
falls. First, false-negative studies may occur in low-grade
chronic osteomyelitis. Second, cellulitis can be difficult
to distinguish from septic arthritis. Third, false-positive
studies can result from normal radiolabeled white cell
uptake in bone marrow around a prosthesis. The combi-
nation of white blood cell and sulfur colloid marrow
scanning is combined to avoid this pitfall. Infection is
diagnosed only in areas of radiolabeled white blood cell
uptake that are negative for marrow activity.
BONE MARROW SCINTIGRAPHY
Bone marrow scintigraphy is not important in current
practice but does have a small number of indications.
The procedure is most commonly performed with
Tc-99m sulfur colloid, which localizes in the reticuloen-
dothelial elements of the red marrow.
In patients with sickle cell anemia,bone marrow imag-
ing demonstrates the extent of marrow expansion into
the extremities (see Fig. 6-46). In most normal subjects,
the marrow is confined to the proximal thirds of the
femurs and humeri.In patients with hemoglobinopathies,

Figure 6-46 Sickle cell anemia: technetium-99m sulfur colloid
bone marrow scan. The intensity has been set to maximize
visualization of marrow. Note the intense uptake in the liver. The
uptake in the bone marrow extends throughout the upper and
lower extremities. The numerous focal defects indicate marrow
infarctions. Based on one examination, new abnormalities cannot
be distinguished from old ones.
Skeletal Scintigraphy 153
Figure 6-47 Osteomyelitis of the right clavicle. Anterior
scintigram in a child show uptake on the right is markedly greater
than in the left clavicle.
the marrow uptake is typically seen throughout the
appendages. Marrow imaging is highly sensitive for
detection of bone marrow infarction and can also define
the extent of involvement. The major limitation is not
being able to distinguish new from chronic infarctions.
Moreover, areas involved by osteomyelitis demonstrate
defects on marrow imaging,so the technique is not useful
in distinguishing infarction from osteomyelitis. In current
practice, marrow imaging is often used in conjunction
with labeled white blood cells to diagnose osteomyelitis.
As noted in the discussion of prosthesis evaluation, in
marrow-bearing areas the combination of increased radio-
labeled white blood cell uptake and absence of Tc-99m
sulfur colloid uptake is specific for osteomyelitis.
BONE MINERAL MEASUREMENT
A number of methods have been developed for quantita-
tive measurement of bone mineral mass. Until recently,
all of the techniques were based on the absorption of
photons in bone, the differential absorption in bone tis-
sue versus soft tissue, and calibration of absorption per-
centage against calcium-containing reference standards.
A number of ultrasound techniques, based on the rate of
sound transmission through bone, are now available.
The simplest technique is single-photon absorptiome-
try (SPA). In this technique, a photon source (typically
154 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-48 Osteomyelitis of the spine. A, Posterior spot view of an adult show intensely increased
uptake in the midlumbar region,involving more than one vertebral level. B, Corresponding radiograph
shows destructive and sclerotic changes involving the L2 vertebral body and adjacent portions of L1 and
L3. The process has involved the intervertebral disks with loss of height in the disk space.

Box 6-14 Three-Phase Skeletal
Scintigraphy: Interpretive
Criteria
Osteomyelitis: arterial hyperemia, progressive focal
skeletal uptake with relative soft tissue clearance; in
children a focal cold area may be seen if osteomyelitis
is associated with infarction
Cellulitis: venous (delayed) hyperemia, persistent soft
tissue activity; no focal skeletal uptake (may have mild
to moderate diffusely increased uptake)
Septic joint: periarticular increased activity on dynamic
and blood pool phases that persists on delayed
images; less commonly the joint structures appear
cold if pressure in the joint causes decreased flow or
infarction
iodine-125) is collimated and scanned across the radius
or calcaneus or both. These sites are selected for the
minimal soft tissue because correction for soft tissue
attenuation is not possible in the SPA technique.
Dual-photon absorptiometry (DPA) again uses a col-
limated photon source that is scanned over the skeletal
part of interest. It is a more flexible technique because
the soft tissue attenuation can be corrected based on
the differential absorption of the beam at different
energies. In DPA, gadolinium-153 with photon energies
of 40–100 keV is typically used. Areas frequently stud-
ied with DPA are the lumbar spine and both the neck
and intertrochanteric region of the femur. The single-
and dual-photon techniques have given way to x-ray-
based approaches in current practice.
Bone densitometry measurements can also be made
with either dedicated x-ray densitometer devices or
special quantitative computed tomography (QCT) algo-
rithms. Single- and dual-energy techniques have been
described for both x-ray and QCT. The advantage of the
x-ray technique is higher photon flux compared with
SPA and DPA instruments. Radiation exposure is essen-
tially identical for dual-energy x-ray densitometry and
DPA. The most versatile and widely used technique in
current practice is dual energy x-ray absorptiometry
( DXA). DXA is the basis for the World Health
Organization criteria for categorizing osteopenia and
osteoporosis.
The main advantage of QCT is the ability to measure
cortical and trabecular bone separately. Dual-energy

Figure 6-49 Osteomyelitis: three-phase study. A, Sequential dynamic images in a middle-aged
man with diabetes and osteomyelitis. Note the intense arterial phase hyperemia.
QCT has the additional advantage over single-energy
QCT of allowing correction for fat in the marrow space.
Both techniques are quite flexible with respect to body
part examined. QCT is an important research tool but is
too expensive for population screening.
Several ultrasound devices are now approved by the
U.S. Food and Drug Administration for measurement of
bone mass. Sound is transmitted faster in dense bone than
in osteopenic bone and the devices are calibrated against
other methods to correlate with bone mass. Application
of the technique is limited to peripheral structures such
as the calcaneus. The low cost, small size, and ease of use
of ultrasound devices make them attractive for popula-
tion screening. However, current data indicate that spine
measurements are necessary to follow the effects of ther-
apy because the spine is the most sensitive structure for
assessing response to drug treatment.
Skeletal Scintigraphy 155
Continued
The main application of bone mineral measurements
is to establish baseline diagnostic measurements in
the evaluation of patients with suspected osteopenia
and osteoporosis and to follow the course of therapy.
Primary osteoporosis has been divided into two sub-
types. Type I or postmenopausal osteoporosis is related
to decreased estrogen secretion after menopause. Type
II or senile osteoporosis is presumably due to age-related
impaired bone metabolism. Risk factors for osteoporosis
include female sex, Caucasian or Asian race, smoking,
chronic alcohol intake, and a positive family history.
Early menopause, long-term treatment with cortico-
steroids, and a number of nutritional disorders includ-
ing malabsorption are also risk factors. Obesity is
protective.
The World Health Organization has established a
classification system for bone mass based on DXA
156 NUCLEAR MEDICINE: THE REQUISITES

Figure 6-49, cont’d B, Blood pool images already show localization in skeletal structures. C,
Delayed static images reveal intense focal accumulation in multiple areas of the great toe and distal
first and second metatarsals.
 Skeletal Scintigraphy 157

Figure 6-50 Cellulitis: three-phase study. A, Sequential images from the flow phase of a diabetic
patient with cellulitis. The ankle region and the visualized portion of the leg show marked
hyperemia. B, Follow-up blood pool images revealed diffusely increased activity in the areas
corresponding to the hyperemia. No focal abnormality was demonstrated on follow-up late-phase
imaging that would suggest osteomyelitis.

measurements of the spine and femoral neck. A meas-

Box 6-15 Lesions That Can Mimic
urement in an individual is compared with the mean and
Osteomyelitis on Three-Phase
standard deviation (SD) for a young control population.
Skeletal Scintigraphy
A reading within 1 SD is considered normal. Osteopenia
is taken as 1–2.5 SD below the control mean and osteo-
Osteoarthritis porosis is defined as 2.5 SD or more below the control
Gout mean. When the standard deviation is reported in this
Fracture
way, it is referred to as the T score.
Stress fracture
The use of bone mineral density has been accelerated
Osteonecrosis (healing)
Charcot’s joint by the availability of new drugs such as alendronate,
Osteotomy a bisphosphonate that localizes in bone and promotes
Reflex sympathetic dystrophy syndrome mineralization. Estrogen is also widely used in post-
menopausal women but is not uniformly well-tolerated,
158 NUCLEAR MEDICINE: THE REQUISITES

and concerns remain about its effects on other diseases

such as breast cancer.

SUGGESTED READING

Collier BD, Fogelman I, Rosenthal L: Skeletal nuclear medicine.

St Louis, Mosby, 1996.
Connolly LP, Strauss J, Conolly SA: Role of skeletal scintigraphy
in evaluating sports injuries in adolescents and young adults.
Nucl Med Ann 171-209, 2003.
Fournier RS, Holder LE: Reflex sympathetic dystrophy: diagnos-
tic controversies. Semin Nucl Med 28:116-123, 1998.
Freeman LM, Blaufox MD: Metabolic bone disease. Semin
Nucl Med 27:195-305, 1997.
Freeman LM, Blaufox MD: Orthopedic nuclear medicine ( Part
I). Semin Nucl Med 27:307-389, 1997.
Freeman LM, Blaufox MD: Orthopedic nuclear medicine (Part
II). Semin Nucl Med 28:1-131, 1998.
Treves ST: Pediatric nuclear medicine. New York, Springer,
1998.

Figure 6-51 Loose hip prosthesis. Anterior and posterior

whole body images in a patient with severe scoliosis and skeletal
metastases from carcinoma of the breast. The patient has a loose
left femoral prosthesis. Uptake is increased at the tip of the
prosthesis and subtly increased in the region of the trochanters,
especially the greater trochanter. Uptake is intensely increased in
the right femoral head because of degenerative arthritic changes.
 7
CHAPTER Hepatobiliary System

Cholescintigraphy Methodology
Radiopharmaceuticals Image Interpretation
Tc-99m IDA Chemistry Normal Hepatic Vascular Anatomy
Kit Preparation Normal Distribution
Mechanism of Uptake and Clearance Diagnostic Criteria
Pharmacokinetics Accuracy
Dosimetry Tc-99m Sulfur Colloid Liver-Spleen Imaging
Patient History Mechanism of Localization and Pharmacokinetics
Patient Preparation Preparation
Methodology Dosimetry
Cholecystokinin Clinical Applications
Physiology Methodology
Sincalide Image Interpretation
Methodology Liver
Normal Cholescintigraphy Spleen Scintigraphy
Blood Flow TC-99m MAA Hepatic Arterial Perfusion
Liver Morphology Scintigraphy
Hepatic Function Mechanism of Localization and Pharmacokinetics
Gallbladder Filling Methodology
Biliary Clearance Clinical Applications
Clinical Applications for Cholescintigraphy Image Interpretation
Acute Cholecystitis
Acute Acalculous Cholecystitis Nuclear medicine has played an important role in
Chronic Cholecystitis liver and spleen imaging for decades, although many
Biliary Duct Obstruction of the radiopharmaceuticals, methodologies and
Choledochal Cyst indications have changed over time. Cholescintigraphy
Biliary Atresia exemplifies the continued importance of functional
Postoperative Biliary Tract and physiological imaging for diagnosis of a variety of
Tc-99m Red Blood Cell liver Scintigraphy hepatobiliary diseases, including acute cholecystitis,
Pathology biliary obstruction, and biliary leak in an era with high
Diagnostic Imaging quality and rapidly improving anatomic imaging modal-
Ultrasonography ities.
Computed Tomography Various other functional scintigraphic liver and spleen
Magnetic Resonance Imaging imaging studies use radiopharmaceuticals with different
Tc-99m-Labeled Red Blood Cell (Tc-99m RBC) Scintigraphy physiological mechanisms of uptake and distribution
Radiopharmaceutical (Table 7-1, Fig. 7-1). These include Tc-99m-labeled red
Mechanism of Localization and Pharmacokinetics blood cells for diagnosis of cavernous hemangioma and
Dosimetry for functional splenic imaging, and Tc-99m MAA to assess

159
160 NUCLEAR MEDICINE: THE REQUISITES

Table 7-1 Liver Radiopharmaceuticals and Clinical CHOLESCINTIGRAPHY

Indications
Cholescintigraphy with Tc-99m iminodiacetic acid (Tc-
Mechanism 99m-IDA) analogs is indicated for diagnosis of various
Radiopharmaceutical of uptake Indication acute and chronic hepatobiliary diseases ( Box 7-1). The
most common indication is for the diagnosis of acute
Tc-99m sulfur colloid Kupffer cell uptake Focal nodular
cholecystitis, but cholescintigraphy also often provides
hyperplasia
Tc-99m iminodiacetic Hepatocyte uptake Cholescintig- unique physiologic information on bile drainage in
acid (HIDA) raphy patients with suspected biliary obstruction and bile leak-
Tc-99m red blood cells RBC labeling/ blood Cavernous age. Dynamic cholescintigraphy is used to evaluate blood
(RBCs) pool distribution hemangioma flow to the liver, hepatic extraction, biliary excretion,
Tc-99m macroaggregated Blood flow, capillary Hepatic
patency of the biliary tract, and gallbladder contraction,
albumin ( MAA) blockage arterial
perfusion all providing important diagnostic information.
Xenon-133 Lipid soluble Focal fatty
tumor
uptake Radiopharmaceuticals
Gallium-67 citrate Lactoferrin transport Tumor/
and iron binding abscess Iodine-131 rose bengal was introduced in 1955 as the
imaging first hepatobiliary imaging agent extracted by hepato-
F-18 fluorodeoxyglucose Glucose metabolism Tumor cytes and cleared through the biliary tract. However, its
( F-18 FDG) imaging
suboptimal imaging characteristics and relatively high
radiation dosimetry limited its clinical utility. I-123
labeled rose bengal was superior in both regards but
never gained widespread use because of limited avail-
ability and, perhaps more importantly, the introduction
in the 1970s of Tc-99m-IDA hepatobiliary imaging radio-
pharmaceuticals.

Tc-99m IDA Chemistry

Tc-99m labeled IDA radiopharmaceuticals were origi-
nally synthesized to be heart-imaging agents, based on
the structural similarities between IDA and lidocaine
molecules (Fig. 7-2). The high hepatic extraction of an
early IDA compound (dimethyl IDA) prompted the
acronym HIDA, for hepatobiliary IDA.

Box 7-1 Cholescintigraphy: Clinical

Indications
Figure 7-1 Anatomy of a liver lobule. Plates of hepatic cells
(hepatocytes and Kupffer cells) are distributed radially around the
central vein. Branches of the portal vein and hepatic artery located Acute cholecystitis
at the periphery of the lobule deliver blood to the sinusoids. Blood Acute acalculous cholecystitis
leaves through the central vein (proximal branch of hepatic veins). Biliary obstruction
Peripherally located bile ducts drain bile canaliculi that course Postcholecystectomy syndrome
between hepatocytes. Sphincter of Oddi dysfunction
Biliary leak
Chronic acalculous cholecystitis
Biliary diversion procedure
hepatic intraarterial perfusion of tumors and liver to
Biliary stent
predict chemotherapeutic drug distribution and effec-
Focal nodular hyperplasia
tiveness. Technetium-99m sulfur colloid (Tc-99m SC) Hepatocellular carcinoma
liver and spleen scintigraphy has a more limited role Enterogastric bile reflux
today than in the past, but important applications Biliary atresia
persist.
 Hepatobiliary System 161

CH3
O
3 2 CH2CH3
Lidocaine 4 1 NH C CH2 N
5 6 CH2CH3

CH3
CH3 H3C
O O
CH2COO OOCH2C
HIDA (dimethyl IDA) NH C CH2 N 99mTc N CH2 C NH
Lidofenin CH2COO OOCH2C
Technescan CH3 H3C
H3C CH3 H3C CH3

CH3 CH
O O
CH2COO OOCH2C
DISIDA (disopropyl IDA) NH C CH2 N 99mTc N CH2 C NH
Disofenin CH2COO OOCH2C
Hepatolite
CH CH

H3C CH3 H3C CH3

CH3 H3C
O O
CH2COO OOCH2C
99mTc
Bromotriethyl IDA H3C NH C CH2 N N CH2 C NH CH3
Mebrofenin CH2COO OOCH2C
Choletec
Br CH3 H3C Br

Biological activity Radioactivity Biological activity

Figure 7-2 Chemical structure of Tc-99m IDA radiopharmaceuticals. Note similarity of Tc-99m
aminoacetic acid analogs (Tc-99m IDAs) to lidocaine. Radioactivity is located centrally (Tc-99m),
bridging two ligand molecules. Iminodiacetate (NCH2COO) attaches to Tc-99m and the acetanilide
analog (IDA) of lidocaine at the periphery carries the biological activity. Substitutions on aromatic
rings differentiate the various Tc-99m IDAs and determine their pharmacokinetics.

Numerous Tc-99m IDA analogs were developed with
different chemical substitutions around the aromatic ring
(see Fig. 7-2). Subsequent analogs offered improvements
in hepatocellular uptake and more rapid blood clearance.
These analogs were known by many acronyms (e.g.,
BIDA, DIDA, EIDA, PIPIDA).
The U.S. Food and Drug Administration (FDA) has
approved three Tc-99m IDA radiopharmaceuticals. The
first agent approved was Tc-99m lidofenin (HIDA); how-
ever, it is no longer in clinical use. The two presently
available radiopharmaceuticals are Tc-99m disofenin
(DISIDA; Hepatolite, DuPont-Merck) and Tc-99m mebro-
fenin (BrIDA; Choletec, E.R. Squibb).
The IDA radiopharmaceuticals are organic anions that
act as bifunctional chelates. The iminodiacetate ( NCH2
COO) attaches at one end to the radiotracer (Tc-99m)
and at the other end to an acetanilide analog of lidocaine,
which determines the biological function (see Fig. 7-2).
Relatively minor structural changes in the phenyl ring (N
substitutions) result in significant alterations in the IDA
pharmacokinetics (Table 7-2).
The final Tc-99m IDA complex exists as a dimer, with
two molecules of the chelating agent (IDA) reacting with
one atom of Tc-99m. This configuration, with Tc-99m
serving as a bridging atom between the two ligand mole-
cules, confers stability to the technetium complex and
determines hepatobiliary excretion.
Kit Preparation
The Tc-99m IDA radiopharmaceuticals are available as
kits that contain the IDA analog and stannous chloride
in lyophilized form. The Tc-99m–IDA complex is
formed by the simple addition of pertechnetate to the
vial. The product is stable for at least 6 hours after
reconstitution.
Mechanism of Uptake and Clearance
Tc-99m IDA radiopharmaceuticals have the same hepato-
cyte uptake, transport, and excretion pathways as biliru-
bin. After intravenous injection, Tc-99m IDA is tightly
162 NUCLEAR MEDICINE: THE REQUISITES

Table 7-2 Normal Pharmacokinetics of Tc-99m

Iminodiacetic Acid Analogs (IDAs)

Hepatic Clearance 2-hr Renal

uptake half-life excretion
Agent (%) (min) (%)

Tc-99m lidofenin (HIDA) 84 42 >14

Tc-99m disofenin (DISIDA) 88 19 <9
Tc-99m mebrofenin (BrIDA) 98 17 <1

bound to protein in the blood, minimizing renal clear-

ance. The radiotracer is transported into the hepatocyte
by a high-capacity, carrier-mediated, anionic clearance
mechanism. After hepatocellular uptake, the tracer is
transported into bile canaliculi by an active membrane
transport system. Tc-99m IDA compounds are stable
in vivo and, in contrast to bilirubin, are excreted in the
original radiochemical form without being conjugated or
undergoing significant metabolism. Because they travel
the same pathway as bilirubin, Tc-99m IDA agents are
subject to competitive inhibition by high levels of serum
bilirubin.
Pharmacokinetics
Once the Tc-99m IDA radiopharmaceutical is excreted
from the hepatocytes and reaches the bile canaliculi,
it follows the physiologic flow of bile. Approximately
two-thirds of bile flow enters the gallbladder via
the cystic duct and the remainder travels through the
common duct and the sphincter of Oddi into the sec-
ond portion of the duodenum (Fig. 7-3). The relative
flow into each is determined by the patency of the bil-
iary ducts, sphincter of Oddi tone, and intraluminal
pressures.
With older Tc-99m IDA compounds (e.g.,Tc-99m lid-
ofenin [HIDA]), serum bilirubin levels greater than
5 mg/dl resulted in poor image quality. However, today’s
radiopharmaceuticals (Tc-99m disofenin and mebro-
fenin) provide diagnostic images with bilirubin levels as
high as 20–30 mg/dl because of their higher extraction
efficiency (see Table 7-2). Mebrofenin has somewhat
greater hepatic uptake and resistance to displacement
than disofenin and is preferable in patients with hepatic
dysfunction.
Poor liver function causes altered biliary pharmacoki-
netics (i.e., delayed uptake and delayed clearance). The
kidneys serve as an alternative route of excretion for IDA
radiopharmaceuticals, normally excreting up to 10% of
the dose (see Table 7-2) with a larger percent as hepatic
function decreases.
Figure 7-3 Physiology of bile flow and pharmacokinetics of
Tc-99m IDA. Bilirubin is transported in the blood bound to
albumin, extracted by the hepatocyte, secreted into the bile
canaliculi, and cleared through the biliary tract into the bowel.
Hepatic uptake and clearance of Tc-99m IDA is similar to bilirubin
except that it is not conjugated or metabolized.
Dosimetry
Tc-99m disofenin and Tc-99m mebrofenin result in simi-
lar radiation absorbed dose to the patient. The highest
estimated radiation dose (target organ) is usually to the
large bowel, approximately 2 cGy (2 rads) (Table 7-3).
The radiation dose to the gallbladder is variable, depend-
ing on whether the gallbladder fills, its ability to contract,
and the length of time before it is stimulated to contract.
Patient History
The patient’s clinical history should be reviewed care-
fully before initiating cholescintigraphy. The nuclear
physician should know the answer to these questions:
What is the clinical question being asked by the referring
physician? Are the symptoms acute or chronic? When
and what did the patient last eat? Has the patient
received any drugs which could affect normal biliary
physiology and thus interpretation of results? Has ultra-
sonography or other imaging been performed, and if so,
what did it show? Has the patient had biliary surgery?
 Hepatobiliary System 163

Table 7-3 Dosimetry for Tc-99m Iminodiacetic acid Box 7-2 Cholescintigraphy: Protocol
(IDA) Radiopharmaceuticals Summary

cGy/185 MBq (rads/5 mCi) PATIENT PREPARATION

Disofenin Mebrofenin Nothing by mouth (NPO) for 4 hours before study.

(Hepatolite) (Choletec) If fasting longer than 24 hours, infuse sincalide.

Liver 0.19 0.24 RADIOPHARMACEUTICAL

Gallbladder 0.60 0.69 Tc-99m mebrofenin or Tc-99m disofenin, 5 mCi
Large intestine 1.90 2.37
intravenous injection:
Urinary bladder 0.46 0.14
Adults: bilirubin <2 mg/dl 5.0 mCi (185 MBq)
Ovaries 0.41 0.51
Testes 0.03 0.03 2–10 mg/dl 7.5 mCi (278 MBq)
Marrow 0.14 0.17 >10 mg/dl 10.0 mCi (370 MBq)
Total body 0.08 0.10 Children: 200 μCi/kg (no less than 1 mCi or 35 MBq)

If the patient has had a biliary diversion procedure, what
is the anatomy? Are there any intra-abdominal tubes or
drains? If so, where are they placed and which tubing
drains each? Should they be open or clamped to answer
the clinical diagnostic question?
INSTRUMENTATION
Camera: large-field-of-view gamma camera
Collimator: low energy, all purpose, parallel hole
Window: 15% over 140-keV photopeak
PATIENT POSITIONING
Supine; upper abdomen in field of view.

Patient Preparation
The patent must not eat for 3–4 hours prior to starting the
study to ensure that the gallbladder is not contracted,
which would prevent radiotracer entry. It is important to
know when the last meal was and if that meal contained
enough fat to cause gallbladder contraction (>10 gm).
If the patient has been fasting for more than 24 hours,
the gallbladder will likely be full of concentrated viscous
bile which can also prevent radiotracer entry. The
patient should receive cholecystokinin (CCK) prior to
the study to empty the gallbladder. Tc-99m IDA must not
be administered until at least 30 minutes after cessation
of the CCK infusion to allow adequate time for gallblad-
der relaxation.
All morphine-related drugs must be withheld for
approximately 6 hours prior to the study because they
can cause a functional partial biliary obstruction that
may be indistinguishable from a true obstruction. In
urgent situations, naloxone (Narcan) has been used prior
to the study to reverse the effect of opiate drugs.
COMPUTER SETUP
1-sec frames × 60, then 1-min frames × 60
IMAGING PROTOCOL
1. Inject Tc-99m IDA intravenously as a rapid infusion
and start computer.
2. At 60 min, acquire right lateral and left anterior
oblique images.
3. If the gallbladder has not filled and acute
cholecystitis is suspected, inject morphine sulfate
(MS) intravenously, 0.04 mg/kg over 1 min (if there is
good biliary duct clearance and biliary-to-bowel
transit).
Acquire study for additional 30 minutes.
Tc-99m IDA reinjection may be necessary if liver
activity has washed out.
4. Perform delayed imaging at 2 to 4 hr if:
a. MS is not administered and gallbladder has not
filled.
b. Other indications (e.g., hepatic insufficiency,
partial common duct obstruction, suspected
biliary leak).

Methodology imaging, morphine sulfate, and CCK are all optional

A standard protocol for cholescintigraphy is summarized maneuvers to be discussed.
in Box 7-2. An initial 60-second flow study (1–3 sec-
onds/frame) is followed by 1-minute frames acquired for
59 minutes. Right lateral and left anterior oblique views Cholecystokinin
are obtained at 60 minutes to aid in confirming or Physiology
excluding gallbladder filling that may otherwise be diffi- CCK is a 33-amino acid polypeptide hormone with
cult to ascertain because of duodenal activity. Delayed numerous physiological effects, most notably gallbladder
164 NUCLEAR MEDICINE: THE REQUISITES
Box 7-3 Gastrointestinal Actions of
Cholecystokinin
Contracts gallbladder
Relaxes sphincter of Oddi
Stimulates intestinal motility
Inhibits gastric emptying
Reduces gastrointestinal sphincter tone
Stimulates hepatic bile secretion
Stimulates pancreatic enzyme secretion
contraction and relaxation of the sphincter of Oddi (Box
7-3). The C-terminal octapeptide of CCK is the physiolog-
ically active portion of the hormone. Sincalide (Kinevac,
Squibb) is the only commercial synthetic form of CCK
available in the United States. It is the C-terminal
octapeptide.
Bile is normally stored and concentrated in the gall-
bladder. The bile is discharged into the intestines in
response to gallbladder contraction and sphincter of
Oddi relaxation. In the small bowel, bile acids play an
important role in fat absorption.
The fat in an ingested meal stimulates CCK secretion
from the mucosa of the proximal small bowel and its sys-
temic release. The serum CCK level rises gradually and
then plateaus (Fig. 7-4). The interaction of CCK with
receptors in the gallbladder wall and sphincter of Oddi
initiates gallbladder contraction and relaxation of the
Figure 7-4 Physiology of gallbladder contraction.After
ingestion of a fatty meal, serum cholecystokinin (CCK) rises to
a peak between 15 and 30 minutes, depending on the type of
meal and rate of gastric emptying. Gallbladder contraction
begins before peak serum CCK. CCK continues to be released
by the duodenum until food empties from the stomach and
proximal bowel.The gallbladder remains contracted until
serum CCK falls below contraction threshold. It then relaxes
and ceases emptying.
sphincter. The serum CCK remains elevated and the gall-
bladder stays contracted until the production of CCK
declines after the meal has cleared from the stomach and
proximal small bowel.
Sincalide
This synthetic commercially available analog of CCK is
commonly used as a pharmacologic intervention in con-
junction with cholescintigraphy. Many of the diagnostic
indications are listed in Box 7-4.
Since the early days of oral cholecystography, both
fatty meals and injectable CCK analogs have been used to
evaluate gallbladder contraction. Although a fatty meal
may be more physiological, this approach has disadvan-
tages. It assumes normal gastric emptying, which can be
quite variable between patients. Slow gastric emptying
will result in a delay in endogenous stimulation of CCK.
Also, the degree of gallbladder contraction depends on
the fat content of the ingested meal.
Intravenous sincalide permits better standardization
and reproducibility than a fatty meal. CCK analogs have
been used with ultrasonography to evaluate gallbladder
contraction. However, cholescintigraphy has proven
more accurate and reproducible. Ultrasonography is
operator-dependent and quantification assumes geomet-
ric assumptions of gallbladder shape. With cholescintig-
raphy, frequent image acquisition that does not require
technologist interaction is routine and quantification of
the gallbladder ejection fraction is volume-based (counts
are proportional to volume) (Box 7-5).
Methodology
When administered during oral cholecystography in the
1970s,it was found that bolus infusions of CCK can cause
spasm of the gallbladder neck in some patients and result
in ineffective gallbladder contraction. Thus it became
common practice in the 1980s with cholescintigraphy to
infuse sincalide intravenously over at least 30 seconds,
Box 7-4 Clinical Indications for Sincalide
Infusion
BEFORE HIDA STUDY
Empty gallbladder in patient fasting longer than 24 hr.
Diagnose sphincter of Oddi dysfunction.
AFTER HIDA STUDY
Differentiate common duct obstruction from functional
causes.
Exclude acute acalculous cholecystitis if gallbladder fills.
Diagnose chronic acalculous cholecystitis.
Confirm or exclude chronic calculous cholecystitis.
 Hepatobiliary System 165

usually 1–3 minutes. The administered dose has varied

Box 7-5 Cholecystokinin (Sincalide) among all users from 0.01–0.04 mg/kg, although the most
Cholescintigraphy: Protocol commonly used is 0.02 mg/kg.
Recent studies have shown that even 1–3 minute
SINCALIDE ADMINISTRATION infusions of 0.01–0.02 mg/kg sincalide result in inef-
Three validated methods with normal values fective contraction of the gallbladder in up to one
1. 60-min infusion: Set up computer acquisition for third of normal subjects (Table 7-4). However, slower
60 1-min frames (128 × 128). infusions of 30–60 minutes result in good contraction
Infuse 0.01 μg/kg of sincalide over 60 min using in the same subjects. The explanation is thought to be
infusion pump. that some persons have spasm of the neck of the gall-
Abnormal GBEF is <40% (Ziessman et al. 2001) bladder with similar to 1–3 minute infusions similar to
or
that seen with a bolus infusion. The spasm is caused
2. 30-min infusion: Set up computer acquisition for 30
1-min frames.
by the supraphysiological nature of the infusion rate
Infuse 0.02 μg/kg of sincalide over 30 min using (Fig. 7-5).
infusion pump. Normal values have never been established for the
Abnormal GBEF is <30% (Ziessman et al. 1992) 3-minute infusion method. The value of 35% as abnor-
or mal was used for many years without substantiating
3. 45-min infusion: Set up computer acquisition for 60 normal data. Recent studies that have attempted to
1-min frames establish normal values have been unsuccessful
Infuse sincalide at dose rate of 0.02 μg/kg/min for because of the wide range of individual response to
45 min using infusion pump this dose rate. However, normal values have been
Calculate GBEF at 60 min established for 30-, 45-, and 60-minute infusions and
Abnormal GBEF is <40% (Yap et al. 1991)
this methodology is recommended (see Table 7-4).
or
Protocols for these three methodologies are described
FATTY MEAL PROTOCOL (see Box 7-5).
1. Set up computer for 60 1-min frames (128 × 128) Furthermore, approximately half of patients receiving
2. Ingest 240 ml (8 oz. can) of Ensure Plus (Abbot a 1- to 3-minute infusion have abdominal cramping and
Laboratories) nausea, which bears no relationship to whether or not
3. Calculate GBEF at 60 min. the patient has hepatobiliary pathology but rather to the
4. Abnormal GBEF is <33% (Ziessman et al. 2003) methodology of infusion. Adverse symptoms are not
seen with slower infusions of 30–60 minutes.
COMPUTER PROCESSING GALLBLADDER EJECTION
FRACTION (GBEF) When clinically indicated (see Box 7-4), sincalide can
1. Select region of interest for the gallbladder and
be infused twice in a patient study because of its short
adjacent liver background. 2.5-minute half-life in serum. For example, sincalide
2. Derive time-activity curve. may be given prior to cholescintigraphy for a patient
3. Calculate GBEF = maximum counts minus minimum fasting for more than 24 hours prior to the study fol-
counts divided by maximum counts, corrected for lowed by another infusion after the 60-minute study to
background. evaluate gallbladder contraction. Caution is indicated in
giving sincalide after a patient has received morphine

Table 7-4 Comparison of 3-Minute and 30- to 60-Minute Sincalide Infusion Methods

Gallbladder Ejection Fraction in Normal Subjects

Study Dose Infusion length GBEF range (%) False positives* Normal values

Ziessman 1992 0.02 mg/kg 3 min 0-100 8/23 (35%) >0%

Ziessman 2001 0.01 mg/kg 3 min 12-74 6/20 (30%) >7%
Ziessman 1992 0.02 mg/kg 30 min 25-97 2/23 (9%) >30%
Ziessman 2001 0.01 mg/kg 60 min 52-88 1/20 (5%) >40%

*GBEF value <35%.

166 NUCLEAR MEDICINE: THE REQUISITES

300

200 100

Plasma Cholecystokinin (pmol/l)

100

Gallbladder EF%
15
50

10

0 30 60 90 0 30 60 90
Time (min) Time (min)
A B
Figure 7-5 Response of serum CCK and gallbladder to one and 60 minute infusions of sincalide.
A, With sincalide infusion of 60 minutes (open circles), the plasma CCK pattern of increase and
plateau is very similar to that seen with fatty meal ingestion (see Fig. 7-4). However, with a one
minute infusion (closed circles), serum CCK has a very rapid upslope and supraphysiological peak,
with rapid return to baseline. B, Gallbladder contraction with a slow infusion (open circles) is both
more prolonged and to a greater degree. (Modified from Hopman et al. Br. Med J 292: 375, 1986, with
permission.)

sulfate because morphine’s pharmacological effect of
4–6 hours may counteract the effect of sincalide.
When sincalide is given before cholescintigraphy,
Tc-99m IDA should not be injected until 30 minutes after
cessation of sincalide infusion to allow time for full gall-
bladder relaxation. When given prior to cholescintigra-
phy, sincalide may alter Tc-99m IDA pharmacokinetics,
causing a delay in biliary-to-bowel transit, discussed in
the Biliary Obstruction section.
When sincalide is not available, a fatty meal may be
administered as an alternative. Normal values will vary
depending on the amount of fat in the meal and the
length of the study. Thus, a meal with an established pro-
tocol and validated normal values should be used (see
Box 7-5).
Normal Cholescintigraphy
Blood Flow
Because of the liver’s predominantly venous origin of
blood flow (75% portal vein and 25% hepatic artery), the
liver is not normally seen during the early arterial blood
flow phase, but rather 6–8 seconds after flow to the
spleen and kidneys (Fig. 7-6).
Early diffusely increased blood flow may be seen
when there is arterialization of the liver’s blood supply
(e.g., with cirrhosis or generalized tumor involvement).
Focally increased liver blood flow occurs with an intra-
hepatic malignant mass or abscess. Increased blood
flow to the region of the gallbladder fossa can be seen
with severe acute cholecystitis. Localized extrahep-
atic increased blood f low may suggest unexpected
pathology.
Liver Morphology
During the early hepatic phase, prior to biliary secretion,
liver size and the presence or absence of intrahepatic
lesions can be assessed. Because the study is usually
acquired dynamically only in the anterior view, this infor-
mation is limited.
Hepatic Function
Liver function can be assessed visually by noting the rapid-
ity of Tc-99m IDA extraction and clearance from blood
pool. Blood pool in the heart should clear by 5–10
 Hepatobiliary System 167

Figure 7-6 Normal technetium-99m IDA studies. A, Top three rows, Two-second blood flow
images. Blood flow to the liver is delayed compared with the spleen and kidneys because of the
liver’s predominantly portal venous blood flow. Bottom, Heart blood pool seen on immediate image
clears over next two frames at 5 and 10 minutes, consistent with good hepatic function.
Continued

minutes (see Fig.7-6). Delayed blood pool and background
clearance is seen with hepatic dysfunction (Fig. 7-7 ).
Hepatic dysfunction may be prehepatic in origin (e.g.,con-
gestive heart failure), hepatic (e.g., hepatitis or cirrhosis),
or posthepatic (e.g.,late common duct obstruction).
Gallbladder Filling
The gallbladder normally begins to fill by 10 minutes and
is usually filled by 30–40 minutes, although 60 minutes
is the standard time interval defined as normal (see Fig.
7-6). The normal size and shape is quite variable.
168 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-6, cont’d B, Five-minute summed images for 60 minutes in a different patient. Right,
left, and common hepatic ducts are seen by 15–20 minutes and common bile duct by 30 minutes.
Biliary-to-bowel clearance is noted at 36 minutes. Gallbladder is visualized early.

Biliary Clearance Acute Cholecystitis

The smaller peripheral biliary ducts are not usually visu-
alized unless enlarged. The left and right hepatic bile
ducts, common hepatic duct, and common bile duct are
usually seen, particularly with frequent image acquisition
and computer cinematic display (see Fig. 7-6). The left
hepatic ducts may be prominent because of the anterior
position of the left lobe.
Although evidence of dilation can be seen on cho-
lescintigraphy, duct size cannot be accurately determined.
The strength of cholescintigraphy is that the functional
patency of normal-sized or enlarged biliary ducts can be
evaluated. The common bile duct is usually seen by 20 min-
utes and clears to less than 50% of peak activity by 60 min-
utes. Transit from the biliary ducts to the small intestine
(biliary-to-bowel transit) normally occurs by 60 minutes.
Clinical Applications for Cholescintigraphy
Some of the more common clinical indications for cho-
lescintigraphy will be discussed (see Box 7-1).
The most frequent indication for Tc-99m IDA cho-
lescintigraphy at most hospitals is to confirm or exclude
clinically suspected acute cholecystitis.
Pathophysiology
Acute cholecystitis is usually initiated by obstruction of
the cystic duct. An impacted calculous (stone) is the
cause in over 95% of cases. Following obstruction, a pro-
gression of sequential histopathological inflammatory
changes in the gallbladder wall ensues beginning with
venous and lymphatic obstruction (Box 7-6), which
results in edema followed by white blood cell infiltration,
hemorrhage, ulceration, necrosis, and finally gangrene
and perforation.
Clinical Presentation
Characteristic clinical symptoms are the acute onset of col-
icky right upper quadrant pain, nausea, and vomiting.
Patients have leukocytosis but usually normal liver function
tests. Even for patients with classic symptoms,typical phys-
ical findings, and suggestive laboratory values, a confirma-
tory imaging study is usually required prior to surgery.
 Hepatobiliary System 169

Figure 7-7 Delayed gallbladder visualization with severe hepatic insufficiency.Very slow blood
pool clearance and poor liver-to-background ratio caused by liver dysfunction. Gallbladder is not
visualized until 12 hours. Last two images are right and left anterior oblique views.

senting in the emergency room with acute abdominal pain

BOX 7-6 Pathophysiology of Acute
Cholecystitis – Sequential
Progression of Events
Cystic duct obstruction
Venous and lymphatic outflow obstruction
Mucosal edema and congestion
Neutrophilic leukocyte infiltration
Hemorrhage and necrosis
Gangrene
Perforation
Ultrasonography
An impacted stone in the cystic duct or neck of the gall-
bladder is diagnostic of acute cholecystitis;however,this is
a rare finding with ultrasonography. Most patients with
acute cholecystitis have cholelithiasis and ultrasonography
has a high sensitivity for detection of gallbladder stones.
However,this finding is not specific for acute cholecystitis.
Asymptomatic gallstones are common and often not
related to the cause of abdominal pain. Of patients pre-
who have gallstones on ultrasonography, less than half are
ultimately diagnosed with acute cholecystitis.
Many of the ultrasonographic findings seen with acute
cholecystitis are nonspecific. Thickening of the gallblad-
der wall and pericholecystic fluid occur with other acute
and chronic diseases. Intramural lucency from edema is
a more specific indicator of acute inflammation. The
“sonographic Murphy’s sign”(localized tenderness in the
region of the gallbladder) in experienced hands is
reported to be diagnostic. However, this observation is
operator-dependent and not always reliable.
The combination of gallstones,intramural lucency,and
the sonographic Murphy’s sign makes the diagnosis of
acute cholecystitis very likely. However, most patients
with acute cholecystitis do not have all these findings.
Investigations that have directly compared sonography
with cholescintigraphy for the diagnosis of acute chole-
cystitis have found cholescintigraphy to be superior
(Table 7-5).
An advantage of ultrasonography is that it can detect
other findings and diseases that may be causing the
patient’s symptoms, such as common duct dilation
170 NUCLEAR MEDICINE: THE REQUISITES

caused by biliary obstruction, pancreatic and liver

Table 7-5 Acute Cholecystitis: Accuracy of tumors, renal stones, pulmonary consolidation, and pleu-
Cholescintigraphy and Ultrasonography ral effusion. Practically, cholescintigraphy and ultra-
sonography provide complementary information.
Sensitivity/Specificity % Cholescintigraphy
Cholescintigraphy is considered the study of choice to
Cholescin- Ultraso-
Study n tigraphy nography confirm the diagnosis of acute cholecystitis because it
reveals the underlying pathophysiology (i.e., cystic duct
Stadalnik 1978 120 100 / 100 70 / 93 obstruction, manifested by nonfilling of the gallbladder).
Weissmann 1979 90 98 / 100
Scintigraphic Diagnosis Nonfilling of the gall-
Freitas 1980 186 97 / 87
Suarez 1980 62 98 / 100 bladder by 60 minutes after Tc-99m IDA injection is
Szalabick 1980 271 100 / 98 abnormal but not specific for acute cholecystitis
Weissmann 1981 296 95 / 99 (Fig. 7-8). Chronic cholecystitis is a common cause for
Zeman 1981 200 98 / 82 67 / 82 nonvisualization at 60 minutes but the gallbladder fills on
Worthen 1981 113 95 / 100 67 / 100
delayed imaging. Nonvisualization of the gallbladder by
Mauro 1982 95 100 / 94
Ralls 1982 59 86 / 84 86 / 90 3–4 hour delayed imaging is diagnostic of cystic duct
Freitas 1982 195 98 / 90 60 / 81 obstruction and, in the appropriate clinical setting, acute
Samuels 1983 194 97 / 93 97 / 64 cholecystitis. Morphine sulfate, discussed later, is now
often used as an alternative to delayed imaging to confirm
the diagnosis of acute cholecystitis (Fig.7-9).
At times it may be difficult to differentiate gallblad-
der filling from activity in overlapping duodenum and

Figure 7-8 Delayed gallbladder visualization with chronic cholecystitis. At 60 minutes, the
gallbladder is not definitely visualized. Focal activity is seen just lateral to the proximal common
duct. However, the next image in the left anterior oblique view confirms that the focal activity is due
to duodenal activity. Gallbladder visualization occurs at 90 minutes.
 Hepatobiliary System 171

Figure 7-9 Morphine-augmented cholescintigraphy. A, Patient with clinically suspected acute

cholecystitis. Tc-99m IDA images for 60 minutes show good visualization of common hepatic and
common bile ducts, biliary clearance into the duodenum, but no gallbladder visualization. Morphine
sulfate (MS) is given. The gallbladder is not visualized, confirming diagnosis of acute cholecystitis by
90 minutes.
Continued

common duct. Right lateral and left anterior oblique
(LAO) views at 60 minutes can be very helpful to con-
firm or rule out gallbladder filling (Fig. 7-10). In the
right lateral projection, the gallbladder lies anteriorly.
In the LAO projection, the gallbladder, an anterior
structure, moves to the patient’s right and the common
duct and duodenum, more posterior structures, move
to the patient’s left. Upright imaging and ingestion of
water are techniques that can be used to clear duode-
nal activity.
Accuracy Many studies have reported on the
sensitivity and specificity of cholescintigraphy for
making the diagnosis of acute cholecystitis (Table 7-5).
More than two-thirds of these investigations reported the
sensitivity to be greater than 95% and the specificity
greater than 90%. Morphine-augmented cholescinti-
graphy has accuracy similar to the delayed imaging
method (Table 7-6).
Although some false positives (nonfilling of the gall-
bladder in patients without acute cholecystitis) occur, the
specificity can be maximized by anticipating situations
associated with an increased incidence of a false-positive
study (Box 7-7) and by using proper methodology.
The optimal time window for performing cho-
lescintigraphy (i.e., after fasting for 3–4 hours but not
more than 24 hours) has been discussed. Patients with
hepatic insufficiency have delayed uptake and clearance
of Tc-99m IDA, which may result in delayed and unpre-
dictable gallbladder visualization because of the altered
pharmacokinetics. Delayed imaging up to 24 hours may
be necessary (see Fig. 7-7).
The most common cause for a false-positive study is
chronic cholecystitis. Less than 1% of patients with
chronic cholecystitis have a totally obstructed cystic
duct caused by chronic inflammation. Another 5%
have delayed gallbladder visualization because of par-
tial cystic duct obstruction (see Fig 7-8). However,
some will have a functional obstruction with nonfill-
ing of the gallbladder caused by a noncontracting
gallbladder full of viscous bile. Delayed imaging or
morphine administration minimizes this potential
cause for false-positive cholescintigraphy but does not
eliminate it.
Another important group at increased risk for false-
positive cholescintigraphy for acute cholecystitis is sick
hospitalized patients with concurrent serious illness.
172 NUCLEAR MEDICINE: THE REQUISITES
Figure 7-9, cont’d B, A second patient shows no gallbladder filling at 60 minutes (upper left),
so MS was given. Gallbladder filling begins within 5 minutes and is definite by 10 minutes
(arrowhead). Acute cholecystitis is ruled out.
They often have had no oral intake for days or weeks and
are receiving hyperalimentation (see Box 7-7).
False-negative studies (gallbladder filling in a patient
with acute cholecystitis) are very uncommon. This can
occur with incomplete obstruction of the cystic duct.
Obstruction of the distal cystic duct can result in proxi-
mal dilation which might be misinterpreted as a small
gallbladder, called the cystic duct sign or nubbin sign. As
discussed later, acute acalculous cholecystitis has
a higher false-negative rate than acute calculous chole-
cystitis. Very rarely, morphine administration may con-
vert a true positive to a false negative in a patient with
acute cholecystitis.
Morphine Augmentation The 3- to 4-hour time
period often required for cholescintigraphy using the
delayed imaging method to diagnose acute cholecystitis
is not optimal for an acutely ill patient, who requires
prompt diagnosis and therapy. As an alternative to
delayed imaging, morphine sulfate is now often routinely
used to shorten the duration of cholescintigraphy.
Morphine sulfate increases intraluminal biliary pres-
sure by constricting the sphincter of Oddi, which causes
preferential flow of bile to and through the cystic duct, if
it is patent. The dose of morphine required to constrict
the sphincter of Oddi is less than that required for pain
relief.
Morphine sulfate, 0.04 mg/kg, is infused intravenously
when the gallbladder does not fill by 60 minutes. With
cystic duct patency, the gallbladder will then begin to fill
by 5–10 minutes after morphine infusion and filling will
be complete by 20–30 minutes. Thus, the entire Tc-99m
IDA study requires only 90 minutes (see Fig. 7-9).
Morphine should not be given if there are scinti-
graphic findings suggestive of biliary obstruction, such as
poor clearance into the bowel (delayed biliary-to-bowel
transit) and/or significant retention of radiotracer in bil-
iary ducts (<50% clearance from peak). Intravenous mor-
phine produces a functional partial common duct
obstruction that cannot be differentiated on scintigra-
phy from a pathologic obstruction caused by stone or
 Hepatobiliary System 173

Figure 7-10 Differentiating gallbladder from common duct and duodenum. Separating the
gallbladder from the common bile duct and duodenum is difficult (middle row) before obtaining
right lateral (RL) and LAO views. The LAO view delineates the common bile duct, but no biliary-to-
bowel clearance has occurred and therefore there is no duodenal activity. In the LAO view, the
gallbladder moves to the right (anteriorly), and the common bile duct and duodenum move to the
left (posteriorly).

Table 7-6 Accuracy of Morphine-Augmented Box 7-7 Causes for False Positive
Cholescintigraphy Cholescintigraphy for Acute
Cholecystitis
Study Sensitivity (%) Specificity (%)
Fasting <4 hr
Choy 96 100
Kim 100 100
Fasting >24 hr
Keslar 100 83 Concurrent severe illness
Vasquez 100 85 Chronic cholecystitis
Fig 94 69* Hepatic insufficiency
Flancbaum 97 95 Hyperalimentation
Fink-Bennett 95 96 Alcoholism (?)
Kistler 93 78* Pancreatitis (?)

*High percentage of patients with concurrent illness and chronic cholecystitis.

stricture. Thus, the diagnosis of biliary obstruction can-
not be made once morphine is given. If biliary obstruc-
tion has not been excluded by 60 minutes, morphine
should not be administered. Delayed images should then
be obtained for up to 3–4 hours.
Rim Sign Increased hepatic uptake adjacent to the
gallbladder fossa (rim sign) is seen on cholescintigraphy in
approximately 25–35% of patients with acute cholecystitis
(Fig.7-11). It may be seen throughout the initial 60-minute
study; however, it is easier to discern as Tc-99m IDA clears
from normal hepatocytes in the rest of the liver.
The importance of the rim sign is twofold. First, it is
a very specific scintigraphic finding for acute cholecysti-
tis. This sign becomes diagnostically useful to confirm
that a patient at increased risk for false-positive cho-
lescintigraphy (see Box 7-7) (e.g., a sick hospitalized
174 NUCLEAR MEDICINE: THE REQUISITES

Box 7-8 Clinical Conditions Associated

with Acute Acalculous
Cholecystitis

Postoperative
Multiple trauma
Extensive burns
Shock
Acquired immunodeficiency syndrome
Mechanical ventilation
Vasculitis
Multiple transfusions

As the name suggests, acute acalculous cholecystitis

Figure 7-11 Rim sign. Increased liver uptake in the region of
the gallbladder fossa. Normal biliary-to-bowel transit, but no
gallbladder visualization at 60 minutes. Delayed imaging showed
no gallbladder filling. The rim sign is specific for acute
cholecystitis and is associated with increased complications.
patient with concurrent serious illness) does indeed
have acute cholecystitis. The rim sign confirms that non-
filling of the gallbladder is a true-positive, not a false-posi-
tive study.
Second, the rim sign identifies patients with acute
cholecystitis who have more severe disease and are at
increased risk for complications (e.g., gallbladder perfo-
ration and gangrene). Even if they do not have these
complications, patients with the rim sign tend to be at
a later stage of the pathophysiological spectrum of dis-
ease (e.g., hemorrhage and necrosis) rather than edema
and white blood cell infiltration (see Box 7-6).
The pathophysiological mechanism for the rim sign is
explained by severe local inflammation in the liver adjacent
to the gallbladder. In severe acute cholecystitis, the gall-
bladder inflammatory process may spread directly to the
adjacent normal liver, which results in increased blood
flow to that region. The high extraction efficiency of the
liver for Tc-99m IDA results in local increased uptake
compared to adjacent noninflamed liver. Regional
delayed clearance may be secondary to inflammatory
edema with obstruction of biliary canaliculi.
is acute cholecystitis without stones, either in the gall-
bladder or obstructing the cystic duct. However, in the
majority of cases, there is cystic duct obstruction. The
obstruction is caused by inflammatory debris,inspissated
bile, and/or local edema, perhaps aggravated by dehydra-
tion. However, there is a substantial minority of patients
with acute acalculous cholecystitis who do not have cys-
tic duct obstruction but rather have direct inflammation
of the gallbladder wall as a result of systemic infection,
ischemia, or toxemia.
There has long been a concern that the accuracy for
cholescintigraphy for the acalculous variety of acute
cholecystitis would be less than for acute calculous
cholecystitis, particularly in those patients without cystic
duct obstruction. False negative cholescintigraphy (gall-
bladder filling) might occur because of the lack of cystic
duct obstruction. Data over the years has been conflict-
ing but does suggest a lower sensitivity for the diagnosis
of acute acalculous cholecystitis compared to the calcu-
lous form of the disease (Table 7-7).
If a false-negative study (filling of the gallbladder) is
suspected in a patient with a high clinical suspicion for
acute acalculous cholecystitis, sincalide may be helpful.
An acutely inflamed gallbladder would not be expected
Table 7-7 Acute Acalculous Cholecystitis: Accuracy
of Cholescintigraphy
Study n Sensitivity (%) Specificity (%)

Shuman 1984 19 68
Acute Acalculous Cholecystitis Weissmann 1983 15 93
The acalculous form of cholecystitis is a life-threatening Mirvis 1986 19 90 61
disease that occurs in seriously ill hospitalized patients, Swayne 1986 49 93
Ramanna 1984 11 100
usually in the intensive care unit (Box 7-8). Because of its
Flancbaum 1995 16 75 100
high mortality (30%) and morbidity (55%), early diagno- Prevot 1999 14 64 100
sis is imperative. However, because of concomitant seri- Mariat 2000 12 67 100
ous illness, diagnosis is often delayed.
 Hepatobiliary System 175

to contract normally. Thus, good contraction would
exclude the disease. However, poor contraction is not
specific. The patient may have an underlying chronic
cholecystitis or be on drugs (Box 7-9) or have a disease
process (Box 7-10) that inhibits gallbladder emptying.
No adverse affects have been reported with the adminis-
tration of sincalide to patients with suspected acute acal-
culous cholecystitis.
A radiolabeled leukocyte study could be used to con-
firm the diagnosis. In-111 leukocytes are preferable to
Tc-99m HMPAO leukocytes because the latter radiophar-
maceutical clears through the biliary tract. Early imaging
at 1–2 hours before biliary clearance may obviate this
problem. The disadvantage of In-111-leukocytes is the
usual imaging time at 24 hours, although earlier imaging
might be useful (e.g., at 4 hours). If positive, the diagno-
sis is confirmed; if negative, repeat imaging at 24 hours
may be indicated.
Chronic Cholecystitis
Symptomatic chronic cholecystitis presents with recur-
rent colicky right upper quadrant pain. The clinical diag-
Box 7-9 Drugs Associated with Poor
Gallbladder Contraction
Morphine
Atropine
Nifedipine (calcium channel blocking agent)
Indomethacin
Progesterone
Oral contraceptives
Octreotide
Theophylline
Benzodiazepine
Phentolamine (alpha-adrenergic blocking agent)
nosis is typically made by detection of gallstones on ultra-
sonography. Laparoscopic cholecystectomy is the usual
treatment. Histopathologically, the gallbladder wall is
infiltrated with chronic inflammatory cells and fibrotic.
Routine 60-minute cholescintigraphy is usually normal.
Less than 5% of patients have abnormal findings (e.g.,
nonfilling or more commonly delayed filling of the gall-
bladder) or delayed biliary-to-bowel transit. These find-
ings are neither sensitive nor specific for chronic
cholecystitis.
Cholelithiasis is common but often asymptomatic.
Patients with asymptomatic gallstones usually have nor-
mal gallbladder contraction. However, patients with
symptomatic chronic cholecystitis typically have poor
gallbladder contraction, as demonstrated by a low gall-
bladder ejection fraction (GBEF) when stimulated with
sincalide or a fatty meal. Thus,sincalide cholescintigraphy
has been used to differentiate patients with symptomatic
chronic cholecystitis from those with asymptomatic gall-
stones and chronic recurrent abdominal pain from other
causes. However, the primary role of sincalide cho-
lescintigraphy in chronic cholecystitis is to diagnose the
acalculous form of chronic cholecystitis.
Chronic Acalculous Cholecystitis
The acalculous variety of chronic cholecystitis is uncom-
mon, occurring in approximately 5% of patients with
symptomatic chronic cholecystitis. It is clinically and
histopathologically indistinguishable from chronic calcu-
lous cholecystitis, except for the lack of gallstones. Over
the years, it has been referred to by various other names
including acalculous biliary disease and cystic duct syn-
drome (Box 7-11). Because the symptoms may be similar
to other disease processes, a noninvasive imaging
method to preoperatively confirm the suspected diagno-
sis is desirable.
Numerous investigations between 1980 and 1990
reported that sincalide cholescintigraphy can confirm
the diagnosis of chronic acalculous cholecystitis. Poor
gallbladder contraction after sincalide infusion can

BOX 7-10 Diseases Associated with Poor

Gallbladder Contraction
BOX 7-11 Synonyms for Recurrent Pain
Diabetes mellitus Syndromes of Biliary Origin
Sickle cell disease
Irritable bowel syndrome
Truncal vagotomy CHRONIC ACALCULOUS CHOLECYSTITIS
Pancreatic insufficiency Acalculous biliary disease
Crohn’s disease Gallbladder spasm
Celiac disease Cystic duct syndrome
Achalasia
Dyspeptic syndrome SPHINCTER OF ODDI DYSFUNCTION
Obesity Papillary stenosis
Cirrhosis Biliary spasm
Pregnancy Biliary dyskinesia
176 NUCLEAR MEDICINE: THE REQUISITES
preoperatively predict postcholecystectomy sympto-
matic relief and histopathological evidence of chronic
cholecystitis (Fig. 7-12); a normal GBEF excludes the
disease.
The most scientifically rigorous investigation is
a prospective randomized study by Yap and colleagues.
The investigation found the positive predictive value of
sincalide cholescintigraphy is greater than 90%. Other
studies have found similar results but were mostly retro-
spective. This is also the only clinical study that estab-
lished its own normal values based on their slow method
of infusion (see Box 7-5). The published studies that have
confirmed a high accuracy for sincalide cholescintigraphy
investigated a selected group of patients who have been
extensively worked up to exclude other diseases and were
followed for many months or years, thus allowing time for
other diseases to present. Thus, the patients referred for
sincalide cholescintigraphy had a relatively high likelihood
of having chronic cholecystitis. Sincalide cholescintigra-
phy confirmed the clinically suspected diagnosis. The
accuracy of this test in a group of patients with a lower
likelihood of disease might not be expected to be as high.
Sincalide cholescintigraphy should be performed after
an appropriate clinical workup. It is best done on an out-
patient basis while the patient is asymptomatic and not
acutely sick or hospitalized because other diseases and
therapeutic drugs may adversely affect gallbladder func-
tion (see Boxes 7-11 and 7-12).
Figure 7-12 Chronic acalculous cholecystitis. Extremely poor contraction of the gallbladder is
evident after sincalide infusion (GBEF, 20%). Biliary-to-bowel transit is delayed, a nonspecific finding
that may be seen with chronic cholecystitis.
Normal (or abnormal) values for GBEF depend on the
method of infusing sincalide. Normal values have not
been established for a 3-minute infusion. That method
should not be used. For a 30-minute infusion, less than
30% is abnormal; for a 45–60 minute infusion, less than
40% is abnormal (see Box 7-5 and Table 7-4).
A persistent misconception is that reproduction of the
patient’s pain with CCK infusion is diagnostic of chronic
acalculous cholecystitis. CCK has a variety of effects on
the gastrointestinal tract (see Box 7-3), one being that it
stimulates intestinal motility, which can cause cramping
pain. The pain of patients with irritable bowel syndrome
can be aggravated by CCK infusion. Up to 50% of normal
controls receiving sincalide over 1–3 minutes complain
of abdominal cramps. However, those receiving the same
total dose over at least 30 minutes do not experience
pain, whether or not they have chronic cholecystitis.
BOX 7-12 Sequential Pathophysiology of
High-Grade Biliary Obstruction
1. Complete or near complete biliary obstruction
2. Increased intrabiliary pressure
3. Decreased bile flow
4. Ductal dilation
5. Biliary cirrhosis

Biliary Duct Obstruction
Obstruction of the bile ducts is often suspected clinically
by the patient’s symptoms and elevated serum liver func-
tion tests (e.g., direct serum bilirubin, alkaline phos-
phatase) and confirmed by anatomical imaging methods
(e.g., ultrasonography and magnetic resonance cholan-
giopancreatography [MRCP]). Dilation of biliary ducts is
diagnostic of obstruction in patients who have not had
prior obstruction or biliary surgery.
However, biliary obstruction may occur in the
absence of hyperbilirubinemia or jaundice, which is usu-
ally a late manifestation. Obstruction does not always
result in dilation of the bile ducts. Furthermore, dilation
can occur in the absence of obstruction. In these latter
two situations, cholescintigraphy can play a particularly
important diagnostic role.
Cancerous and noncancerous causes of biliary tract
obstruction usually present differently (Table 7-8).
Malignancy (e.g., pancreatic or biliary duct cancer) typi-
cally causes high-grade obstruction with painless jaun-
dice, whereas choledocholithiasis usually produces
either severe acute pain with complete obstruction or
intermittent pain of low-grade obstruction. Pancreatitis
tends to cause only mild partial obstruction. High-grade
biliary obstruction may present with Charcot’s triad of
fever, chills, and hyperbilirubinemia.
High-Grade Biliary Obstruction
Pathophysiology The sequence of pathophysio-
logical events in high-grade obstruction progresses in
a predictable manner ( Box 7-12), although the time
course may vary depending on the rapidity of onset, the
degree of obstruction, and its etiology. High-grade
obstruction causes increased intraductal pressure. The
resulting backpressure markedly reduces bile flow
and causes biliary duct dilation. With persistence of
obstruction,there is increased hepatocellular permeability
and ultimately hepatocellular damage culminating in
biliary cirrhosis.
Table 7-8 Biliary Obstruction: Clinical
Presentations
Degree of Obstruction Usual Etiology
High-grade—complete obstruction
Painless jaundice Malignancy
Acute severe abdominal pain Choledocholithiasis
Charcot’s Triad (fever, chills, Biliary obstruction
hyperbilirubinemia)
Low-grade—partial obstruction
Recurrent biliary colic Choledocholithiasis
No jaundice or liver function Choledocholithiasis
abnormalities
Hepatobiliary System 177
Hyperbilirubinemia is a late manifestation of obstruc-
tion. Elevation of the serum alkaline phosphatase occurs
earlier in the natural history and is more specific. Ductal
dilation, the anatomical imaging sine qua non of biliary
obstruction, may not become evident until 24–72 hours
after the initiating event. Cholescintigraphy can confirm
the diagnosis before dilation occurs by depicting the
physiology, a reduction in bile flow (Fig. 7-13).
Partial Biliary Obstruction
The pathophysiology of partial obstruction is less well
understood. Ducts often do not dilate with low-grade or
intermittent biliary obstruction, probably because of the
lower intraductal pressure and the intermittent nature of
the process. The etiology is usually choledocholithiasis.
In some cases, ductal dilation may be restricted by edema
and scarring.
Noninvasive Imaging Oftentimes the imaging
diagnosis of partial biliary obstruction is made with
ultrasonography and MRCP. However, in patients without
jaundice, anatomical imaging may be unrewarding. In
these cases, cholescintigraphy can help determine the
need for a more invasive workup, such as endoscopic
retrograde cholangiopancreatography (ERCP) or
percutaneous cholangiography.
Ultrasonography The diagnosis of obstruction is
made by noting dilation of the biliary tree. Dilation is
most common with long-standing obstruction, particu-
larly when secondary to malignant etiologies. Although
the degree of dilation is not directly proportional to the
serum bilirubin, the largest, most dilated ducts tend to
occur in deeply jaundiced patients. An obstructing mass
may be detected with ultrasonography. Biliary duct
stones are rarely seen.
Magnetic Resonance Cholangiopancreatography
MRCP is now widely used for diagnosis of biliary obstruc-
tion. The rapid T2-weighted scans viewed in the coronal
plane mimic the appearance of contrast cholangiography.
It is superior to ultrasonography for detecting stones too
small to result in ductal dilatation. Its diagnostic accuracy
for detection of a variety of benign and malignant
processes approaches ERCP or percutaneous transhep-
atic cholangiography. However, it is still an anatomic
imaging methodology and cannot evaluate bile flow.
Discordance between anatomical imaging and func-
tional imaging with cholescintigraphy is not uncommon
in high-grade or low-grade obstruction. For example, duc-
tal dilation may not be seen, whereas cholescintigraphy
demonstrates abnormal delayed bile flow. Functional
abnormalities may precede morphologically evident dis-
ease. In other patients with dilated ducts from prior
exploration or chronic passage of stones, cholescintigra-
phy may show normal bile flow and no obstruction.
Cholescintigraphy Cholescintigraphy is particularly
useful in those patients with suspected biliary obstruc-
tion who have normal-sized biliary ducts or in patients
178 NUCLEAR MEDICINE: THE REQUISITES
Figure 7-13 High-grade biliary obstruction. Good hepatic uptake, but no secretion into biliary
ducts or gallbladder. The elevated backpressure prevents tracer from entering the biliary system.
who have had prior biliary obstruction and have persist-
ently dilated ducts. The scintigraphic pattern of com-
plete or high-grade obstruction of recent onset is an
“Aunt Minnie,”that is, good hepatic extraction and uptake
of Tc-99m IDA but no excretion into the biliary tree (per-
sistent hepatogram) (see Fig. 7-13). With complete
obstruction, delayed imaging for up to 24 hours may
show no change. With less complete but high-grade
obstruction, there can be slow excretion into biliary
ducts on delayed imaging at 2–24 hours.
The longer the obstruction goes on, the more likely
there will be associated evidence of hepatic dysfunction
(delayed hepatic uptake, prolonged blood pool), which
can make differentiation of primary common bile duct
obstruction from parenchymal dysfunction more difficult.
With partial common duct obstruction,the scintigraphic
findings are quite different than those seen with high-grade
obstruction (Fig. 7-14). Prompt hepatic uptake and secre-
tion into biliary ducts is typically seen. The two most impor-
tant cholescintigraphic findings are delayed transit into the
bowel (delayed biliary-to-bowel transit) and delayed clear-
ance of radiotracer from biliary ducts (Box 7-13).
However, delayed biliary-to-bowel transit is not a sensi-
tive indicator of obstruction (Box 7-14). Up to 50% of
patients with partial obstruction have biliary-to-bowel tran-
sit by 60 minutes. Therefore, it is particularly important to
judge whether there has been good clearance of bile tracer
from the biliary ducts by 60 minutes. Generally,at least 50%
clearance from peak is expected. If not, delayed imaging at
2 hours, or alternatively, sincalide infusion can be given.
Subsequent biliary-to-bowel transit rules out obstruction.
Segmental obstruction will show similar scintigraphic find-
ings as described,but in a regional liver pattern.
Less common and specific findings include filling
defects within distal biliary ducts. This finding is most
likely to be detected on the cinematic display as biliary
duct filling begins. Persistent ductal segmental narrow-
ing with proximal retention of activity is strongly sugges-
tive of partial obstruction. However, these findings are
really anatomical, and the real value of cholescintigraphy
is the ability to analyze biliary physiology. Thus, signifi-
cant retention of Tc-99m IDA within biliary ducts at 60
minutes, with or without biliary-to-bowel transit, should
raise the suspicion of partial obstruction.
In patients who have had previous obstruction or bil-
iary tract surgery, biliary ducts often remain permanently
dilated. Thus, anatomical imaging methods of evaluation
are not helpful. In patients who have recurrent symp-
toms, cholescintigraphy can differentiate obstructive
from nonobstructive dilation. With nonobstructed dila-
tion, there should be good ductal clearance and normal
biliary-to-bowel transit.
Delayed biliary-to-bowel transit has various causes
other than partial biliary obstruction (Box 7-14). Delayed
transit is seen in patients who have received sincalide
prior to the study for the purpose of emptying the gall-
bladder (Fig. 7-15). As the prokinetic effect of sincalide
dissipates, the gallbladder relaxes. The resulting negative
intraluminal gallbladder pressure causes bile to flow
preferentially towards the gallbladder rather than
through the common duct and sphincter of Oddi.
Morphine-related drugs can produce a functional
obstruction with delayed biliary duct clearance and bil-
iary-to-bowel transit. Delayed transit may also be seen in
some patients with chronic cholecystitis. Finally, one
report found that 20% of patients without biliary disease
may have delayed biliary-to-bowel transit, sometimes
referred to as a hypertonic sphincter of Oddi (Fig. 7-16).
Functional causes of delayed biliary-to-bowel transit
can be confirmed by having the patient change position,
 Hepatobiliary System 179

Figure 7-14 Partial common duct obstruction. A, The common bile duct appears increasingly
prominent. Refluxed bile is seen in the left hepatic duct. No biliary-to-bowel transit occurs by 60
minutes. B, Sincalide infusion with sequential images. No gallbladder contraction and minimal
biliary-to-bowel transit occur.The pattern is consistent with partial common duct obstruction.

walk around, and by obtaining delayed images at 2 hours.
With obstruction, there would be persistent retention of
activity within biliary ducts. Alternatively, sincalide can
be given. The use of sincalide is a more reproducible
method. In either case, persistent pooling in the biliary
ducts (particularly in the common duct) after sincalide
infusion is consistent with partial obstruction, whereas
clearance rules out obstruction.
Accuracy For high-grade obstruction, the sensitivity
for diagnosis approaches 100% and the specificity is
greater than 95%. Rarely, cholestatic jaundice (e.g., drug-
induced [erythromycin, chlorpromazine]), may have sim-
ilar findings. For low-grade or intermittent obstruction,
the sensitivity and specificity have been reported to be
95% and 85%, respectively.
Summary Noninvasive anatomic imaging proce-
dures such as ultrasound and CT are the initial screen-
ing procedures in patients with surgical jaundice.
MRCP has become an important anatomic diagnostic
method. However, jaundice occurs relatively late in the
natural history of obstruction. Scintigraphy is useful in
patients whose symptoms are suggestive of obstruction
but who are not jaundiced and have minimal liver func-
tion abnormalities. Scintigraphy may show low-grade

BOX 7-13 Scintigraphic Diagnosis of

Partial Biliary Obstruction Box 7-14 Causes of Delayed Biliary-to-
bowel Transit
Delayed biliary-to-bowel transit beyond 60 minutes
Poor biliary duct clearance at 60 minutes Biliary obstruction
No further biliary duct clearance on delayed imaging at Sincalide administration prior to cholescintigraphy
120 minutes Opiate drugs
No biliary duct clearance with sincalide administration Chronic cholecystitis
between 60 and 120 minutes Normal variation (hypertonic sphincter of Oddi)
180 NUCLEAR MEDICINE: THE REQUISITES
Figure 7-15 Delayed biliary-to-bowel transit due to sincalide pretreatment. Gallbladder fills by 30
minutes, common duct is seen at 60 minutes, but no bowel clearance is seen. Delayed images show
intestinal clearance starting at 90 minutes and decreased activity in the common duct by 2 hours.
obstruction or segmental intrahepatic obstruction.
Cholescintigraphy is also useful in distinguishing
abdominal pain caused by acute cholecystitis from that
secondary to biliary obstruction. It may diagnose the
presence of both problems. Patients who have had
prior obstruction or common duct exploration and per-
sistently dilated atonic biliary ducts are prime candi-
dates for cholescintigraphy.
Choledochal Cyst
A choledochal cyst is a congenital dilation of the biliary
system. It is not a true cyst but rather a saccular or
fusiform biliary duct dilation. Often it involves the com-
mon hepatic and common bile duct but it may occur any-
where within the biliary system, usually extrahepatic but
rarely intrahepatic (Caroli’s disease), and it may be multi-
focal (Fig. 7-17).
A choledochal cyst usually presents clinically in young
children as biliary obstruction, pancreatitis, or cholangi-
tis. However, they may be asymptomatic and found inci-
dentally. On rare occasion, they will first present in
adulthood. Ultrasonography or CT may reveal a cystic
structure but often cannot ascertain whether the struc-
ture connects with the biliary tract.
Cholescintigraphy can confirm that the cystic structure
connects to the biliary system. Tc-99m IDA tracer will fill
the choledochal cyst and often have very prolonged reten-
tion within it, depending on its size and the degree of
obstruction. Delayed images are often required (Fig.7-18).
Biliary Atresia
This neonatal disease of unknown etiology is character-
ized by inflammatory sclerosis and obliteration of extra-
hepatic and intrahepatic bile ducts. Untreated, it leads
to cirrhosis and death within the first years of life.
Treatment involves surgery, a palliative hepatic portoen-
terostomy (Kasai procedure) performed in the neonatal
period and subsequent liver transplantation.
Biliary atresia must be differentiated from neonatal
hepatitis and cholestasis. Clinical features of biliary atre-
sia include jaundice after full-term birth that lasts longer
than 2 weeks, hepatomegaly, and hyperbilirubinemia.
Early diagnosis of biliary atresia is critical because sur-
gery must be performed within the first 60 days of life
before irreversible liver failure ensues.
Infants with idiopathic neonatal hepatitis or cholesta-
sis are often preterm and/or small for gestational age.
The most common causes include Alagille syndrome
(arteriohepatic dysplasia), progressive familial intra-
hepatic cholestasis, alpha-1 antitrypsin deficiency, and
cystic fibrosis.
Cholescintigraphy has long been used to help differ-
entiate biliary atresia from other causes of neonatal jaun-
dice. The congenitally atretic bile ducts produce a picture
 Hepatobiliary System 181

Figure 7-16 Delayed biliary-to-bowel transit in normal subject. Top three rows, Sequential
images acquired over 60 minutes. Gallbladder fills. Biliary ducts are visualized, but no biliary-to-
bowel transit is seen at 60 minutes. Lower two rows, Sincalide is infused over 30 minutes.
Gallbladder contracts (GBEF, 51%), and biliary-to-bowel transit is seen due to concomitant relaxation
of sphincter of Oddi. Arrowhead, Mild gastric reflux. This is a normal subject with a hypertonic
sphincter of Oddi.

of high-grade obstruction (Fig. 7-19). Patients should be
pretreated with phenobarbital (5 mg/kg/day for 5 days)
before cholescintigraphy to maximize sensitivity for
detection. The drug activates liver excretory enzymes.
Ideally, the serum phenobarbital level should be checked
prior to the study to ensure a therapeutic serum level of
the drug.
The usual administered dose of Tc-99m IDA is
200 mCi/kg with 1 mCi being the minimal dose because
24-hour images are required. No biliary clearance into
the bowel by 24 hours is consistent with biliary atresia.
Patients with nonobstructive causes of neonatal jaundice
usually have biliary clearance into the bowel during the
first 24 hours after injection of Tc-99m IDA. False-positive
182 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-17 Schematic drawings of various types of choledochal cysts. Type I: Cystic dilation of
an extrahepatic duct (most common); Type II: sac or diverticulum opening from the common bile
duct;Type III:choledochocele, located within the duodenal wall; Type IV A: involving intrahepatic
and extrahepatic biliary ducts; Type IV B dilatation of multiple segments confined to extrahepatic
biliary ducts; Type V: multiple intrahepatic ducts (Caroli’s disease).

studies (no biliary-to-bowel clearance) occasionally
occur in patients with severe forms of parenchymal liver
disease. When the clinical diagnosis is uncertain, a repeat
study may be indicated.
Postoperative Biliary Tract
Cholescintigraphy provides diagnostic information in
postoperative patients with suspected complications from
biliary tract surgery (e.g., laparoscopic or open cholecys-
tectomy, biliary duct surgery, gallstone lithotripsy, and bil-
iary enteric anastomoses). Cholescintigraphy can be used
in the differential diagnosis of the postcholecystectomy
syndrome. Posttherapeutic evaluation with Tc-99m IDA
can be useful in patients treated for obstruction with
papillotomy/sphincterotomy or biliary stents (i.e., to con-
firm patency or diagnose restenosis) (Fig.7-20).
Postcholecystectomy Pain Syndrome
Recurrent abdominal pain of hepatobiliary origin after
cholecystectomy may be caused by a retained or recur-
rent biliary duct stone, postoperative inflammatory stric-
ture, or sphincter of Oddi dysfunction (see Box 7-15).
Rarely, a cystic duct remnant acts like a small gallbladder
and produces symptoms identical to those of acute or
chronic cholecystitis.
Partial Biliary Obstruction
Partial biliary obstruction caused by retained or recur-
rent common duct stones and inflammatory fibrosis
are the most common causes of postcholecystectomy
 Hepatobiliary System 183

Figure 7-18 Choledochal cyst. 25-year-old with abdominal pain. Sonography noted a cystic
structure adjacent to the common hepatic duct; however, a definite connection to the biliary system
could not be ascertained. A, Tc-99 IDA images acquired at 90 minutes after the liver had cleared
show filling of choledochal cyst in the region of the common hepatic duct (arrowhead ). CD,
Common duct; GB, gallbladder, D, duodenum. B, Cholangiogram confirmed the diagnosis.

Figure 7-19 Biliary atresia. Four-week-old child. Serum phenobarbital level was in therapeutic
range. A, Images every 10 minutes for 1 hour after injection of Tc-99m mebrofenin shows no biliary-
to-bowel transit.
Continued
184 NUCLEAR MEDICINE: THE REQUISITES
Figure 7-19, cont’d B, Images at 4 (not shown) and 24 hours
show no bowel activity, only renal clearance into the bladder.
Surgery confirmed biliary atresia.
syndrome. The scintigraphic findings described for par-
tial common duct obstruction apply, that is, delayed bil-
iary-to-bowel-transit and delayed clearance of biliary
ducts (Fig. 7-21). Because the gallbladder is not present
to act as a pressure release reservoir for bile, ductal
dynamics directly predict the adequacy of biliary
drainage. Diagnostic scintigraphic findings are those of
Figure 7-20 Patent biliary stent. Common duct stent was placed to relieve obstruction from a
malignant tumor.Tc-99m IDA study confirms patency of the stent. Note the hepatic mass in the liver
dome.
persistent activity within the common duct and delayed
biliary-to-bowel transit.
Sphincter of Oddi Dysfunction
Sphincter of Oddi dysfunction (SOD) is a partial biliary
obstruction at the level of the sphincter of Oddi, not
caused by stones or stricture. It occurs in up to 14% of
postcholecystectomy patients and presents as intermit-
tent abdominal pain and transient liver function abnor-
malities. Therapy is often sphincterotomy for a fixed
( papillary stenosis), whereas a functional and reversible
obstruction (biliary dyskinesia) may respond to drugs
(e.g., nifedipine) and sphincter toxins (e.g., Botox).
ERCP has been used to exclude a stone or fibrosis as
the cause for the partial biliary obstruction. Delayed
drainage of contrast material beyond 45 minutes seen
during ERCP is consistent with the diagnosis but non-
specific. MRCP has largely replaced ERCP. Sphincter of
Oddi manometry is the gold standard used by gastroen-
terologists. An elevated sphincter pressure (>40 mm
Hg) is considered diagnostic. However, this technique is
invasive, not widely available, technically difficult, and
prone to interpretative errors. Further-more, medica-
tions given during ERCP can affect the results, repro-
ducibility is a concern, and pancreatitis a serious
complication. Thus a noninvasive alternative would be
preferable.
Cholescintigraphy permits physiological assessment
of duct drainage. Because this entity is a partial biliary
obstruction. Scintigraphic findings are delayed biliary
duct clearance at 60 minutes and no further clearance
between 1–2 hours. Early studies suggested that image
analysis is diagnostic in the majority of cases.

Box 7-15 Causes of Postcholecystectomy
Pain Syndrome
Retained or recurrent choledocholithiasis
Inflammatory stricture
Sphincter of Oddi dysfunction
Nonhepatobiliary origin
Cystic duct remnant (obstrmucted or inflamed)
Various quantitative methods have been used to fur-
ther improve on image analysis (e.g., bile duct t ⁄ , percent
1
2
emptying at specific time intervals). Results have been
mixed. Sincalide has been advocated as a pharmacologic
intervention to improve diagnostic accuracy by increas-
ing bile flow (see Box 7-3) and stressing the capacity of
the biliary ducts, thus revealing more subtle abnormali-
ties that might not otherwise be seen. The reported
accuracy of the various methodologies has varied with
no general consensus on the best method. One tech-
nique that incorporates image analysis and semiquantita-
tive parameters using sincalide infusion is described
(Box 7-16, Fig. 7-22).
It is stated that a characteristic finding of SOD is a para-
doxical response of the sphincter of Oddi to CCK (e.g.,
contraction rather than relaxation) and that this finding
might be used diagnostically to confirm the disease.
However,this paradoxical response was initially described
in animal studies administering CCK as a bolus. It was occa-
sionally reported at manometry in clinical studies but again
Figure 7-21 Biliary stricture causing partial obstruction. Images acquired at 5, 10, 20, 40, and 60
minutes. Common hepatic and bile ducts are dilated above an abrupt distal cutoff; however, there is
normal biliary-to-bowel transit.The patient had a prior cholecystectomy.
Hepatobiliary System 185
the drug was given as a rapid infusion. It has never been
shown that when given at physiological dose rates a para-
doxical response of the sphincter of Oddi actually occurs.
Biliary Leaks
In most cases, bile leaks occur after cholecystectomy or
other biliary tract surgery. Sometimes obstruction can be
the cause. The laparoscopic method has become the pro-
cedure of choice for elective cholecystectomy, however,
it is associated with a somewhat higher rate of bile duct
injury than open cholecystectomy.
Although ultrasonography and CT can detect fluid col-
lections, the type and origin of the collection is often
uncertain. Cholescintigraphy can determine whether the
fluid collection is of biliary origin rather than due to
ascites, hypoproteinemia, or other causes. With cho-
lescintigraphy, the approximate location and rate of bil-
iary leakage can be estimated. This information is
important prognostically. Slow bile leaks usually resolve
spontaneously with conservative therapy; however, more
rapid leaks require surgical correction.
Before percutaneous drainage of a biloma, cholescintig-
raphy can help ensure that central biliary obstruction is
not present. With obstruction, it is unlikely that bile leak-
age can be effectively treated by percutaneous drainage
without addressing the underlying cause of obstruction.
Follow-up cholescintigraphy can be used to confirm reso-
lution or persistence of the leakage.
Scintigraphically, bile leakage often manifests initially
as a progressively increasing collection of radiotracer in
the region of the gallbladder fossa or hepatic hilum. The
activity may progressively spread into the subdiaphrag-
matic space, over the dome of the liver, into the colonic
186 NUCLEAR MEDICINE: THE REQUISITES

gutters,or manifest as free bile in the abdomen (Figs.7-23

BOX 7-16 Sphincter of Oddi Dysfunction and 7-24). Positioning the patient on their right side
Protocol—Sincalide and allows the radiotracer to pool in the gutter. Delayed
Semiquantitative Score imaging may be required to detect a small or slow leak.
Patient repositioning can help confirm its presence (e. g.,
PREPARATION lateral decubitus views). Peritoneal tubing, drains, and
Fasting for 2 hr prior to study collection bags may exhibit accumulation and should be
imaged as well.
COMPUTER SETUP
Biliary Diversion Surgery
1-min frames × 60 Biliary enteric bypass procedures are created in patients
IMAGING PROTOCOL for a wide range of benign and malignant diseases associ-
1. Infuse sincalide 0.02 mg/kg × 10 min (altered from
ated with biliary obstruction. Complications are not
original protocol that was 3-min infusion) uncommon.
2. 15 min after sincalide infusion, inject 5 mCi Tc-99m Ultrasonography has imaging limitations in the pres-
mebrofenin or disofenin intravenously. ence of gas in the anastomotic bowel segment or refluxed
biliary air following surgery. Two thirds of attempts are
COMPUTER PROCESSING ANALYSIS
reported as indeterminate. Biliary dilation may be present
1. Draw regions of interest around liver and common in over 20% of patients even though obstruction has been
duct and derive time-activity curves. adequately relieved by surgery. It may not be possible to
2. Use image analysis in conjunction with time-activity
reach the biliary tract using ERCP if a long Roux-en-Y loop
curves.
3. Time-activity curves will help determine time to
has been created as part of the anastomosis. MRCP has
hepatic peak and % CBD emptying. high accuracy in visualizing the postoperative stricture
and detecting obstruction when new dilation is present.
SCINTIGRAPHIC SCORING However, cholescintigraphy is the only noninvasive
1. Peak liver uptake Score
method that can distinguish obstructed dilated ducts from
a. Less than 10 min 0 those that are chronically dilated but not obstructed.
b. 10 min or greater 1
Scintigraphy is well-suited to diagnose bile leakage,
2. Time of biliary visualization functional patency of the anastomosis, or recurrent
a. Less than 15 min 0 obstruction. It is important to be aware of the postopera-
b. Greater than 15 min 1 tive anatomy of the patient being imaged. Although biliary
scintigraphy is useful in a variety of anastomoses (choledo-
3. Prominence of biliary tract
choduodenostomy, cholecystoduodenostomy, and chole-
a. Not prominent 0 cystojejunostomy), it is particularly valuable in patients
b. Prominence of major extrahepatic ducts 1
with choledochojejunostomy or intrahepatic cholangioje-
c. Prominence of major intrahepatic ducts 2
junostomy. Choledochojejunostomy is a direct anastomo-
4. Bowel visualization sis of the extrahepatic portion of the common bile or
a. Less than 15 min 0 common hepatic duct to a Roux-en-Y jejunal loop. The
b. 15–30 min 1 intrahepatic cholangiojejunostomy requires direct anasto-
c. Greater than 30 min 2 mosis between small bowel and intrahepatic ducts.
5. CBD emptying With scintigraphy, intestinal excretion by 1 hour with
or without ductal dilatation is consistent with function-
a. More than 50% 0
b. Less than 50% 1
ally patent. Intestinal excretion greater than 1 hour is
c. No change 2 suggestive of partial obstruction. However, retention of
d. Increasing activity 3 activity in bile ducts is a more reliable indicator.
Persistent or worsening biliary duct retention between
6. CBD-to-liver intensity ratio
1 and 2 hours is fairly specific for obstruction. Stasis with
a. CBD 60 min ≤ liver 60 min 0 minimal intestinal excretion and pooling in the region of
b. CBD 60 min > liver 60 min but less than the biliary enteric anastomosis may be seen at 1 hour.
liver 15 min 1 This may be positional and can be confirmed and
c. CBD 60 min higher than liver 60 min and
resolved by imaging the patient upright. Persistent non-
equal to liver 15 min 2
d. CBD 60 min higher than both liver 60 min
visualization of the biliary system and intestine suggests
and liver 15 min 3 complete biliary duct obstruction.
Other Gastrointestinal Surgical Anastomoses
TOTAL SCORE Cholescintigraphy can provide functional information
about other surgical procedures involving the gastroin-
Score of >5 is consistent with sphincter of Oddi dysfunction (Sostre et al. 1992).
CBD, common bile duct.
testinal tract (e.g., Billroth I and II,Whipple resection). In
 Hepatobiliary System 187

Figure 7-22 Sphincter of Oddi dysfunction. Above, 2-min sequential images demonstrate
delayed clearance from the common duct. Right, Regions-of-interest are draw for the common duct
and liver. Below, Time-activity curves of the common duct and liver. This patient had a scintigraphic
score of 7, positive for sphincter of Oddi dysfunction. See Box 7-16 for scoring method.

Billroth II anastomoses,Tc-99m IDA can evaluate afferent
loop patency. The afferent loop should fill readily in an
antegrade direction from the common duct. There is nor-
mally progressive accumulation of activity within the
loop which becomes abnormal when the filling persists
for more than 2 hours.
Liver Transplants
The major role for cholescintigraphy is to aid in the
detection of the postoperative complications of bile
leaks and obstruction. The findings of rejection on cho-
lescintigraphy are nonspecific signs of liver dysfunction.
Liver biopsy is necessary to make the diagnosis.
Trauma
Posttraumatic problems that must be clinically differenti-
ated include hepatic laceration, hematoma, bile duct tran-
section, extrahepatic biliary leakage, intrahepatic biloma
formation,and perforation of the gallbladder. CT and ultra-
sonography can diagnose liver parenchymal injury. Only
biliary scintigraphy can demonstrate communication
between the biliary tree and space-occupying lesions that
represent biloma formation.
Bile leakage is common after penetrating and blunt
trauma. It may initially be occult and detected only after
clinical deterioration or discharge of bilious material
from surgical bed drains. Cholescintigraphy can be used
to differentiate between rapid and slow bile leakage,
which can help the surgeon decide whether interven-
tion or watchful waiting is indicated and assess resolu-
tion of leakage in patients treated conservatively.
Primary Benign and Malignant Tumors
Liver tumors that contain hepatocytes would be
expected to take up Tc-99m IDA, thus cholescintigraphy
can be helpful in the differential diagnosis of primary
benign and malignant hepatic tumors (e.g., focal nodular
hyperplasia, hepatic adenoma, and hepatocellular carci-
noma) (Table 7-9).
Focal Nodular Hyperplasia and Hepatic Adenoma
The natural history and therapy of these two benign
188 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-23 Post-cholecystectomy biliary leak. CT scan 4 days after surgery showed
intraabdominal fluid. HIDA study, ordered to determine if the fluid collection was of biliary origin,
confirms a biliary leak. Sequential images are from 60 minutes (left) to 6 hours (right) after
injection. Bile extravasates into the portal region, over the liver dome, and into the colic gutter.

Figure 7-24 Posttransplant biliary leak. Abdominal pain and ascites postoperatively. Sequential
images show prompt intraperitoneal leak emanating from the region of the extrahepatic biliary
ducts. At surgery, an occluded biliary stent was found with active bile leakage.
 Hepatobiliary System 189

is seen in more than 90% of patients with focal nodular

Table 7-9 Differential Diagnosis of Primary hyperplasia. Poor clearance presumably is caused by
Hepatic Tumors with Technetium-99m abnormal biliary canaliculi. The overall accuracy is
HIDA
higher than the traditional Tc-99m sulfur colloid method
used to diagnose focal nodular hyperplasia, which shows
Lesion Flow Uptake Clearance uptake in about two-thirds of patients (see section on
Focal Nodular Hyperplasia Under Tc-99m Sulfur
Focal nodular Increased Immediate Delayed
hyperplasia Colloid ). Hepatic adenomas consist exclusively of hepa-
Hepatic adenoma Normal None — tocytes;therefore it is surprising and not understood why
Hepatocellular Increased Delayed Delayed it rarely exhibits uptake on cholescintigraphy.
carcinoma Hepatocellular Carcinoma Tc-99m IDA cho-
lescintigraphy also demonstrates characteristic findings for
hepatocellular carcinoma (hepatoma). The malignant
hepatocytes are hypofunctional compared to normal liver.
tumors are quite different. Focal nodular hyperplasia is During the first hour of cholescintigraphy,no uptake within
usually asymptomatic, often discovered incidentally, and the lesion (cold defect) is seen. Delayed imaging at
requires no specific therapy, whereas hepatic adenomas 2–4 hours often shows “filling in,” or continuing uptake
are often symptomatic and may cause serious hemor- within the tumor and concomitant clearing of adjacent
rhage that can be life-threatening. Adenomas have normal liver (Fig. 7-26). This pattern is very specific for
a strong association with oral contraceptives;focal nodular hepatoma. However, poorly differentiated hepatomas may
hyperplasia does not. not fill in on delayed imaging. Tc-99m IDA uptake may be
Focal nodular hyperplasia contains all hepatic cell seen at sites of distant hepatocellular metastases (e.g.,lung).
types (i.e., hepatocytes, Kupffer cells, and bile canaliculi). Enterogastric Bile Reflux
Characteristic findings seen with cholescintigraphy are Alkaline gastritis occurs secondary to enterogastric
increased blood flow, prompt hepatic uptake, but reflux, seen most commonly after gastric resection sur-
delayed clearance (Fig. 7-25). This characteristic pattern gery. Symptoms are similar to those of acid-related

Figure 7-25 Focal nodular hyperplasia. Sequential images every 5 minutes show early uptake by
tumor in the dome of the liver (arrowhead) that persists throughout the 60-minute study as the
normal liver clears the tracer. Focal uptake persisted on delayed images acquired at 3 hours.
190 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-26 Hepatocellular carcinoma. A, CT shows a large mass in the posterior aspect of the
right lobe (arrowhead). B, Left, Tc-99m HIDA posterior view acquired at 5 minutes shows a cold
defect in the same mass as seen on CT (arrowhead). Right, Posterior view. Two-hour HIDA images
show increased uptake within the lesion (arrowhead) and good washout of the remainder of the
liver. Surgery confirmed hepatocellular carcinoma.

disease. Cholescintigraphy can demonstrate the bile
reflux (Fig. 7-27). Some bile reflux is not uncommonly
seen in normal subjects on routine cholescintigraphy,
particularly if morphine sulfate or CCK has been
administered. However, the greater the quantity and the
more persistent the reflux, the more likely that the
reflux is related to the patient’s symptoms. Quantitative
methods for estimating the amount of reflux have been
described.
Liver hemangiomas are usually asymptomatic.They are
discovered incidentally on CT or ultrasonography during
the clinical workup or staging of a patient with a known
primary malignancy or during evaluation of unrelated
abdominal symptoms or disease. They require no spe-
cific therapy but must be differentiated from other, more
serious liver tumors.

TC-99M RED BLOOD CELL LIVER
SCINTIGRAPHY
Cavernous hemangiomas are the most common benign
tumor of the liver and the second most common hepatic
tumor, exceeded in incidence only by liver metastases.
Pathology
Cavernous hemangiomas of the liver have abnormally
dilated, endothelium-lined vascular channels of varying
sizes separated by fibrous septa. Histopathologically, they
are not related to capillary hemangiomas, angiodysplasia,
or infantile hemangioendotheliomas. Ten percent of these
cavernous hemangiomas are multiple. Lesions larger than
4 cm are often called giant cavernous hemangiomas.

Figure 7-27 Enterogastric reflux. Sixty minutes after injection
of Tc-99m IDA, reflux of labeled bile into the stomach is seen. Bile
gastritis was confirmed at endoscopy.
Diagnostic Imaging
Noninvasive diagnosis of cavernous hemangioma of the
liver is important because it can obviate the need for
biopsy, which on occasion can result in hemorrhage-
associated morbidity and even mortality.
Ultrasonography
The typical sonographic pattern for hemangioma (i.e.,
a homogeneous, hyperechoic mass with well-defined
margins and posterior acoustical enhancement) is nei-
ther sensitive nor specific for the diagnosis of cavernous
hemangioma.
Computed Tomography
Strict CT criteria for hemangioma include relative
hypoattenuation before intravenous contrast injection,
early peripheral enhancement during the rapid bolus
dynamic phase, progressive opacification toward the
center of the lesion, and complete isodense fill-in, usually
by 30 minutes after contrast administration. Frequently,
not all criteria are satisfied. When these criteria are used
to maximize specificity, the sensitivity of CT is only 55%;
less strict criteria result in a high false-positive rate.
Accuracy is even poorer with multiple hemangiomas.
Magnetic Resonance Imaging
Cavernous hemangiomas have a characteristic appearance
on MRI,with high signal intensity on T2-weighted spin-echo
images (light bulb sign). Gadolinium contrast shows find-
ings similar to those seen with CT with contrast. Although
MRI is much more accurate than CT, various benign and
malignant tumors may give false-positive results. MRI is par-
Hepatobiliary System 191
ticularly helpful in the diagnosis of small lesions and those
adjacent to major vessels or vascular organs.
Tc-99m-Labeled Red Blood Cell ( Tc-99m RBC)
Scintigraphy
With modern multiheaded SPECT cameras,the radionuclide
imaging technique has good sensitivity and an exceedingly
low false positive rate. Although not as sensitive for detec-
tion of small hemangiomas as MRI,it is more specific
Radiopharmaceutical
Labeling the patient’s erythrocytes with Tc-99m pertech-
netate is done by the same methodology as discussed in
Chapter 11 in the section on Gastrointestinal Bleeding,
Radiolabeling with Tc-99m RBCs (see Box 7-9). The
in vitro kit method is now the preferred approach because
of its high labeling efficiency and ease of preparation.
Mechanism of Localization and
Pharmacokinetics
After injection, the Tc-99m-labeled red blood cells (RBCs)
are distributed within the blood pool of the liver. The
labeled cells require time to exchange and equilibrate
within the large, relatively stagnant, nonlabeled blood
pool of the hemangioma (Fig. 7-28). This equilibration
time varies from 30–120 minutes. Thus, early images will
typically show the cavernous hemangioma to be cold but
delayed images will demonstrate increased uptake.
Dosimetry
The total body radiation absorbed dose is approxi-
mately 0.4 cGy (rads). The target organ is the heart
Heart
Spleen
Liver
Figure 7-28 Tc-99m erythrocyte pharmacokinetics in liver
hemangioma. Left, Immediately after injection the hemangioma is
“cold.”Blood pool activity within the liver is greater than activity
within the hemangioma.Time is required for the radiolabeled RBCs
to equilibrate with the unlabeled RBCs in the blood pool volume
of the hemangioma. Middle, As the Tc-99m-labeled RBCs
increasingly enter the hemangioma and mix with the unlabeled
cells, activity in the hemangioma becomes equal to normal liver.
Right, When fully equilibrated (60–120 minutes), activity within
the hemangioma exceeds that in the surrounding liver and is equal
to activity in the heart and spleen.
192 NUCLEAR MEDICINE: THE REQUISITES

wall, which receives 1.2 cGy (rads), followed by the

spleen and kidney (Table 7-10). The bladder dose Box 7-17 Tc-99m Red Blood Cell Liver
depends on the method of labeling and the degree of Hemangioma Scintigraphy:
free Tc-99m pertechnetate. Protocol Summary

PATIENT PREPARATION
Methodology
None
A combined three-phase planar and single-photon emis-
sion computed tomography (SPECT) technique is used RADIOPHARMACEUTICAL
(Box 7-17). SPECT is mandatory for state-of-the-art Tc- Tc-99m pertechnetate, 25 mCi, labeled to RBCs (in vitro
99m RBC hepatic scintigraphy. Planar flow and early kit method)
blood pool images are not necessary for diagnosis, Inject intravenously; bolus injection for flow images
although they show characteristic findings.
INSTRUMENTATION
Camera: large-field-of-view gamma with SPECT
Image Interpretation capability
Image interpretation should be performed with CT or Energy window: 15% centered over 140-keV photopeak
ultrasonography available for direct anatomical correla- Collimator: low energy, high resolution, parallel hole
tion to ensure proper identification of the abnormality in IMAGE ACQUISITION
question. Planar Imaging
1. Blood flow: 1-sec frames for 60 sec on computer and
Normal Hepatic Vascular Anatomy
2-sec film images.
The liver has a complex vascular system (Figs. 7-1 and 2. Immediate images: acquire 750k–1000k count planar
7-29). It receives approximately two-thirds of its blood image in same projection and other views as
supply from the portal vein and only one third from the necessary to best visualize lesion(s).
hepatic artery. The sinusoids act as the capillary bed for 3. Delayed images: acquire 750k–1000k count planar
the liver cells. Blood leaves the liver through the hepatic static images 1–2 hr after injection in multiple
veins, which then empty into the inferior vena cava. The projections (anterior, posterior, lateral, and oblique
caudate lobe is an exception in that it also has a direct views).
connection with the vena cava. Much of this normal vas- Single-Photon Emission Computed Tomography
cular anatomy of the liver is seen with Tc-99m-labeled
1. Position patient supine on imaging table. Raise
RBCs (Figs. 7-30 to 7-34).
patient’s arms above head.
2. Center liver in field of view.
Normal Distribution
3. Rotate camera head around patient to ensure that
Organs with the highest activity per pixel are the heart camera does not come in contact with patient. Liver
and spleen, followed by the kidney. The normal liver has should remain completely in field of view during test
much less blood pool activity. The aorta, inferior vena rotation.
cava, and occasionally the portal vein can be seen with
SPECT Camera Setup
Window 15% window centered over
140-keV Tc-99m photopeak
Setup Step and shoot
Table 7-10 Dosimetry for In Vitro Tc-99m Red Blood Collimator(s) High or ultrahigh resolution
Cell Scintigraphy
Computer Setup
Rads/25 mCi Acquisition Parameters
Target (cGy/925 MBq) Rad/mCi Patient orientation Supine
Rotation Clockwise
Heart wall 1.350 0.054 Matrix 128 × 128 word mode
Bladder wall 1.275 0.051 Image/arc combination 128 images/360°
Spleen 1.025 0.041
Time/frame 10 sec/stop
Blood 0.875 0.035
Liver 0.650 0.026
Reconstruction Parameters
Kidneys 0.625 0.025 Filters Manufacturer specific
Ovaries 0.425 0.017 Personal preference
Testes 0.175 0.007 Attenuation correction Yes
Total body 0.375 0.015 Reformatting Transverse, sagittal, coronal
 Hepatobiliary System 193

Figure 7-29 Vascular anatomy of the liver. Blood supply to the

liver is predominantly from the portal vein (75%) and to a lesser
extent from the hepatic artery (25%). Both enter the liver in the
portal area. Hepatic artery and its branches are not shown here.
Portal vein divides into right and left branches and then
subdivides. Smaller branches with hepatic artery branches and
canaliculi define the periphery of lobules (see Fig. 7-1). Hepatic
veins originate at lobule center (central veins), feeding into right,
middle, and left hepatic veins, which drain into the inferior vena
cava (IVC).

planar imaging. Portal branching vessels and hepatic

veins can be seen with SPECT.

Diagnostic Criteria
With planar imaging, the arterial blood flow to a heman-
gioma is usually normal. Immediate blood pool images
typically have decreased uptake within the hemangioma
compared with adjacent liver, although early increased
uptake is occasionally seen (Figs. 7-31 and 7-32).
On 1- to 2-hour delayed imaging, hemangiomas have
increased activity compared with adjacent liver. The
uptake is usually equal to that of the blood pool of the
heart and spleen. Giant cavernous hemangiomas often
show heterogeneity of uptake on delayed images, with
areas of decreased as well as increased uptake (see Fig.
7-31). These cold regions, often located centrally, are
caused by thrombosis, necrosis, and fibrosis. Benign and
malignant liver tumors, abscesses, cirrhotic nodules, and
cysts have decreased activity compared to normal liver
(see Fig.7-30).
Accuracy
Tc-99m RBC scintigraphy has a very high positive predic-
tive value, approaching 100%, so a positive test is likely to
be a true positive. After more than three decades of clinical
use, very few false-positive studies have been reported,
Figure 7-30 Negative Tc-99m RBC study for hemangioma.
A, CT scan shows a large lesion in right lobe of the liver. B, Planar
Tc-99m RBC scan is cold in the same region and negative for
hemangioma. Metastatic colon cancer was diagnosed.
and it is considerably more specific than CT. Although
a few large hepatomas have been misinterpreted as heman-
giomas, most hepatomas are negative on scintigraphy.
A large series of moderately sized hepatomas found no false
positives. Angiosarcomas are extremely rare but are
a reported cause of false positives. A few other carcino-
mas have been reported to be positive, usually because
they produce local obstruction of sinusoids. Although
extensive fibrosis or thrombosis may rarely result in
a false-negative study,areas of increased uptake are usually
also seen.
The diagnostic sensitivity of Tc-99m RBC imaging
depends primarily on lesion size and the camera system
194 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-31 Giant cavernous hemangioma. Left, Immediate postinjection image shows a large,
relatively photopenic area involving most of the left lobe and a large portion of the right lobe. Some
focal areas of increased uptake are seen. Right, Delayed 1-hour image shows filling of the initial cold
area and increased uptake throughout this large hemangioma that is equal to the heart (H) and
spleen (S).

Figure 7-32 Comparison of planar, single-, and multiheaded SPECT for liver hemangioma.
A, Planar study. Left, Immediate postinjection image shows photopenic region in superolateral
portion of the right lobe with small area of increased uptake. Right, Delayed image at 60 minutes
shows complete filling in, diagnostic of hemangioma.
Continued

used. SPECT is superior to planar imaging because of its
improved contrast resolution (Figs. 7-32 to 7-35). SPECT
is especially useful for the detection of small heman-
giomas, those located centrally in the liver, multiple
hemangiomas (see Figs. 7-32 and 7-35), and those adja-
cent to the heart, kidney, and spleen (see Fig. 7-34). In
seven comparison studies performed between 1987 and
1991, the mean overall sensitivity for planar imaging was
55% and for SPECT was 88% (Table 7-11).
Lesion size and location are critical determinants of
detectability (Table 7-12). Generally, planar imaging
can demonstrate hemangiomas down to about 3 cm in
size. Single-headed SPECT has good sensitivity for
hemangiomas 2 cm and larger, whereas multiheaded

Figure 7-32, cont’d B, SPECT coronal sections. Top, Single-
headed Tc-99m sulfur colloid SPECT coronal section with well-
defined cold defect. Middle, Single-headed camera with
comparable Tc-99m RBC coronal section shows increased uptake
in lesion, consistent with hemangioma. Although contrast
resolution is improved with SPECT, there is no diagnostic
advantage over planar imaging. Bottom, Triple-headed SPECT
shows the hemangioma, as well as another small hemangioma
immediately adjacent (arrow) not seen with single-headed SPECT.
The small hemangioma measured 0.9 cm on CT.
Hepatobiliary System 195
SPECT can detect almost all hemangiomas greater than
1.4 cm and may show lesions as small as 0.5 cm,
although the sensitivity for detection is reduced (see
Table 7-9). With multiheaded SPECT, the increased sen-
sitivity is achieved by using ultra-high-resolution colli-
mators (see Fig. 7-30).
TC-99M SULFUR COLLOID LIVER-
SPLEEN IMAGING
Tc-99m sulfur colloid (SC) was introduced in 1963 and was
the standard clinical method for liver-spleen imaging until
the advent of CT. Although not a frequently requested
study today,there are still occasional clinical indications.
Mechanism of Localization and
Pharmacokinetics
After intravenous injection,Tc-99m SC colloid particles
0.1–1.0 μm in size are extracted from the blood by retic-
uloendothelial cells, the Kupfer cells of the liver (85%),
macrophages of the spleen (10%), and bone marrow
(5%). Tc-99m SC has a single-pass liver extraction effi-
ciency of 95% and a blood clearance half-life of 2–3 minutes.
Uptake is complete by 15 minutes. After phagocytosis, the
Tc-99m SC particles are fixed intracellularly.
Other factors besides extraction efficiency influence
the distribution of colloid particles (e.g., blood flow, par-
ticle size, and disease states). Increased blood flow to
a region of the liver increases the local regional delivery
and extraction of colloid. Larger colloid particles
increase the proportion that will be taken up by the liver.
Conversely, smaller particles will increase distribution to
bone marrow.
Kupffer cells are distributed throughout the liver (see
Fig. 7-1) but make up less than 10% of liver cell mass.
Most liver diseases affect hepatocytes and Kupffer cells
similarly. Disease results in local, diffuse, or heteroge-
neously decreased uptake because of destruction or dis-
placement of normal liver.
With severe diffuse liver disease, a generalized reduc-
tion in hepatic extraction and increased distribution to
the spleen and bone marrow (colloid shift) occurs. With
splenomegaly or immunologically active states, increased
splenic uptake can be seen.
Preparation
Tc-99m SC is available in kit form and requires 15 min-
utes to prepare. Acid is added to a mixture of Tc-99m
pertechnetate and sodium thiosulfate, which is heated in
Figure 7-33 Improved visualization of small hemangioma with
SPECT. A, Left, Immediate postinjection planar image shows neither
decreased nor increased uptake, probably because of small lesion
size. Right, After 1.5-hour delay, planar study shows mildly increased
focal uptake in the liver dome. If lesion had been more central, it may
not have been seen due to overlying activity. B, SPECT coronal (top)
and transverse (bottom) sections are strongly positive for
hemangioma with high target-to-background ratio. Note the heart
(H) and aorta (A).
 Hepatobiliary System 197

Figure 7-34 Negative planar but positive SPECT hemangioma study. A, Liver CT scan shows
lesion of uncertain etiology in the left lobe (arrow). B, Planar anterior (left) and posterior (right)
Tc-99m RBC study is negative, probably because of proximity of lesion to hot spleen. Oblique views
were not helpful. C, SPECT study performed with an old single-headed camera clearly detects the
hemangioma (arrowheads) adjacent to the spleen and the heart’s left ventricle in coronal (right)
and transverse (left) slices.
198 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-35 Increased number of lesions detected with SPECT. A, CT scan shows two lesions in
middle and posterior aspect of the right lobe. B, Only the larger and more posterior lesion is positive
on planar imaging (arrowheads). C, Both lesions are seen with SPECT. The inferior vena cava and
aorta are medial to the hemangiomas.
 Hepatobiliary System 199

a water bath (95–100°F) for 5–10 minutes. Labeling effi-

ciency is greater than 99%. Clinical Applications
The clinical role today for Tc-99m SC is limited to situa-
Dosimetry tions in which it can add functional information not avail-
able from the usual anatomical imaging methods or used
Estimated radiation dose from Tc-99m sulfur colloid is 1.7 as a template for correlating imaging findings with
cGy (rads) to the liver and 1.1 cGy (rads) to the spleen. another radionuclide study and for splenic imaging.
(Table 7-13).

Methodology
No patient preparation is required. Imaging can start
within 20 minutes after radiopharmaceutical injection.
Table 7-11 Planar Imaging vs. SPECT for Tc-99m SPECT should be routine (Box 7-18). When acquired, pla-
RBC Detection of Hemangioma
nar images are obtained in multiple views.

Sensitivity (%)

Study Planar SPECT

Tumah 1987 43 100

Malik 1987 77 100 Box 7-18 Tc-99m SC Liver-Spleen
Brodski 1987 44 78 Scintigraphy: Protocol
Itenzo 1988 88 100 Summary
Brunetti 1988 69 100
Kudo 1989* 42 74
Ziessman 1991* 30 71 PATIENT PREPARATION
Overall 55 88
Study should not be performed immediately after a
barium contrast study since attenuation artifacts may
*Lower sensitivity in later years is caused by the larger number of small lesions.
result.

CONTRAINDICATION
None
Table 7-12 Sensitivity for Hemangioma Detection by
Lesion Size with Multiheaded SPECT RADIOPHARMACEUTICAL
Planar imaging: 4 mCi (148 MBq)
Lesion (cm) Sensitivity (%) SPECT: 6 mCi (296 MBq)
Pediatric patients: 30–50 μg/kg (minimal dose, 300 μCi)
>1.4 100
>1.3 91 INSTRUMENTATION
1.0–2.0 65
Camera: large-field-of-view gamma
0.9–1.3 33
0.5–0.9 20 Window: 15% over 140-keV photopeak
Collimator: parallel hole, low energy, high resolution

IMAGING PROTOCOL
1. Inject technetium-99m sulfur colloid intravenously.
2. Commence imaging 20 min after injection.
Table 7-13 Technetium-99m Sulfur Colloid:
Radiation Dosimetry Planar Imaging
500k–1000k count images in multiple projections (ant.,
Rads/5 mCi (cGy/185 MBq) upright and supine, supine with costal marker, post.,
right and left lateral, ant. and post. oblique).
Organ Normal Diffuse parenchymal disease Upright imaging is preferable when possible to
minimize respiratory excursion, a cause of image
Liver 1.7 0.8
degradation.
Spleen 1.1 2.1
Bone marrow 0.14 0.4 SPECT
Testes 0.006 0.016
Ovaries 0.028 0.06 Acquisition protocol similar to that for technetium-99m
Total body 0.095 0.09 RBC liver scintigraphy (see Box 7-17)
200 NUCLEAR MEDICINE: THE REQUISITES

Abnormal scintigraphic findings include hepato-

Image Interpretation
Interpretation of Tc-99m SC liver-spleen scans requires
an appreciation of the normal variability of liver anatomy,
the effect of extrinsic liver compression by normal and
abnormal structures, and common artifacts (Figs. 7-36
to 7-39). Planar imaging with multiple views has been
used successfully for many years. However, SPECT
improves lesion detection because of its improved con-
trast resolution (Figs. 7-40 and 7-41).
megaly, heterogeneity of distribution, splenomegaly, col-
loid shift, and focal defects. Hepatomegaly is a non-
specific finding that suggests hepatic dysfunction or an
infiltrating process (Box 7-19). Hepatomegaly and het-
erogeneity of uptake may be the only findings in early
lung and breast cancer because these liver metastases are
small and diffusely infiltrating (Fig. 7-42). Colon cancer
metastases to the liver are usually larger and focal at clini-
cal presentation.

Figure 7-36 Normal anatomy of liver. A, Anterior view. IVC, Inferior vena cava. B, Posterior view.

Figure 7-37 Normal anatomical landmarks and interpretative pitfalls for Tc-99m SC liver
scintigraphy. A, Anterior. B, Right lateral. C, Posterior. D, Left lateral. C, Costal rib indentation; GB,
gallbladder fossa; HV, notch from hepatic veins; I, incisura umbilicus (ligamentum teres); K, kidney
impression; L, left lobe; N, notch between right and left lobes; PH, porta hepatis; Q, quadrate lobe; R,
right lobe; V, vertebral spine attenuation; S, spleen.
 Hepatobiliary System 201

Figure 7-38 Normal Tc-99m sulfur colloid liver-spleen scan.Two anterior views, with cold line
marker along costal margin in supine position (top left) and without marker in upright position (top
middle). In sequence, remaining images are right anterior oblique, right lateral, posterior, right
posterior oblique (shallow), left posterior oblique, left lateral, and left anterior oblique.

Figure 7-39 Liver photopenic defect caused by intrahepatic gallbladder. Left, Anterior Tc-99m
sulfur colloid (SC) liver image with photopenic defect in lateral aspect of mid-right lobe. Right,
Tc-99m IDA study performed immediately after sulfur colloid study showed gallbladder filling of
the Tc-99m SC defect.
202 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-40 Normal liver anatomy: SPECT correlated with CT. A, Selected CT transverse
sections. B, Corresponding SPECT slices. GB, Gallbladder fossa; QL, quadrate lobe; PV, portal vein
bifurcation; IVC, inferior vena cava; RL, right lobe; LL, left lobe.
 Hepatobiliary System 203

Figure 7-41 Improved lesion detectability with SPECT. Patient with malignant melanoma
referred for Tc-99m SC study to rule out hepatic metastases. A, Anterior planar images in upright
(left) and supine (right) views. Questionable defect at medial inferior aspect of the left lobe was
thought to be a normal variation. B, Selected SPECT short-axis (top two) and coronal (bottom two)
sections demonstrate well-defined lesion in anterior aspect of the left lobe (arrowheads).
204 NUCLEAR MEDICINE: THE REQUISITES
Box 7-19 Causes of Hepatomegaly
Infiltrative: fatty metamorphosis, alcoholic liver disease,
amyloidosis, Gaucher’s disease,Wilson’s disease,
granulomatous involvement
Congestive: heart failure, hepatic vein thrombosis
Neoplastic: primary and secondary tumors
Infectious: hepatitis, sepsis, malaria
Inflammatory: drugs (e.g., methyldopa, isoniazid)
Miscellaneous: cystic disease
Various methods have been used to estimate liver and
spleen size. Estimates can be made using cobalt hot
markers or lead cold markers at 1- to 2-cm increments.
Generally, the normal liver’s longest vertical and mid-
clavicular line dimensions are 17 and 15 cm, respectively.
Spleen size greater than 14 cm in its longest axis or
greater than 110 cm2 using two perpendicular dimen-
sions is enlarged.
With proper intensity settings, bone marrow uptake
is usually not perceptible. Splenic uptake on the poste-
rior view is normally equal to or less than that of the
liver. Colloid shift (increased splenic uptake) is seen
Figure 7-42 Colon cancer metastases on Tc-99m sulfur colloid (SC) study. A, Anterior and right
lateral views show large metastases in the right and left lobes. B, Good response to chemotherapy.
Extensive liver metastases on initial Tc-99m SC study (left) and definite response to therapy seen on
follow-up 4 months later (right).
with a variety of hepatic diseases (Figs. 7-43 and 7-44).
Quantitative spleen-to-liver count ratios greater than 1.5
are abnormal.
Lung radiocolloid uptake is relatively uncommon.
Causes include improper labeling with excessive alu-
minum causing large-particle clumping and various
pathophysiological processes (e.g., severe liver dysfunc-
tion) (see Fig. 7-44). The lung uptake is probably the
result of activation of normal lung macrophages and
stimulation of RES cell migration from other parts of the
body to the lung.
Liver
Decreased Uptake
Most benign and malignant lesions of the liver produce
“cold” or “photopenic” defects on Tc-99m SC liver imag-
ing (Box 7-20; see Figs. 7-41 and 7-42). Radiation therapy
produces characteristic rectangular port-shaped hepatic
defect. Diffusely decreased uptake is usually caused by
hepatocellular disease, although early infiltrating tumor
involvement appears similar.
Ancillary radionuclide tests have been used in con-
junction with Tc-99m SC to add specificity to the scinti-
graphic diagnosis (e.g., In-111 leukocytes for infection
and Tc-99m RBCs for hemangioma). Xenon-133,
 Hepatobiliary System 205

a fat-soluble agent, exhibits increased uptake in focal
fatty tumors and in generalized fatty metamorphosis of
the liver.
Increased Uptake
Increased hepatic uptake on Tc-99m SC imaging is
uncommon but quite characteristic for specific diag-
noses (Box 7-21).
Superior Vena Cava Obstruction Collateral tho-
racic and abdominal wall vessels communicate with the
recanalized umbilical vein delivering Tc-99m SC via
the left portal vein to a small volume of tissue, usually
in the region of the quadrate lobe, producing a “hot spot”
(Fig. 7-45). This collateral blood flow has a relatively
increased concentration of colloid compared with blood
delivered to the bulk of the liver after systemic mixing.
Injection in the lower extremity rather than the upper
extremity results in a normal scan.
Focal Nodular Hyperplasia Focal nodular hyper-
plasia may have increased uptake because of both the
vascular nature of this benign tumor and the increased
density of functioning Kupffer cells (Fig. 7-46). Focal
nodular hyperplasia can often be confirmed with Tc-99m
SC. Because this tumor has hepatocytes, bile ducts, and
Kupffer cells, normal or increased colloid uptake is seen
in two thirds of patients. One third appears cold for
uncertain reasons. Hepatic adenoma (hepatocytes only)
is always cold.
Budd-Chiari Syndrome Budd-Chiari syndrome
(hepatic vein thrombosis) often has relatively more
uptake in the caudate lobe than the remainder of the
liver (Fig. 7-47). The impaired venous drainage of the
majority of the liver results in poor hepatic function. The
caudate lobe retains good function as a result of its direct
venous drainage into the inferior vena cava.
Alcoholic Liver Disease
Many diseases affect the liver diffusely (Boxes 7-19 and
7-22). Alcoholic liver disease is the most common cause
seen in fatty infiltration, acute alcoholic hepatitis, and

Figure 7-43 Hepatic parenchymal disease on Tc-99m SC study.A, Hyperpigmentation and biopsy-
proven hemochromatosis in 52-year-old man.Anterior (left) and posterior (right) views show small right
lobe,hypertrophied left lobe,large spleen,and colloid shift.B, Severe cirrhosis.Anterior view shows very
small liver with poor uptake,enlarged spleen,and prominent colloid shift to the marrow and spleen.
206 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-44 Tc-99m SC lung uptake due to fatty metamorphosis of the liver of pregnancy. A,
During the patient’s acute illness the liver-spleen scan showed increased lung uptake, colloid shift to
the marrow and spleen, and inhomogeneous liver uptake. B, Follow-up study after patient clinically
recovered. Tc-99m SC liver-spleen scan returned to normal.

Box 7-20 Causes of Focal Liver Defects Box 7-21 Causes of Increased Focal
Uptake on Tc-99m SC Imaging
Cyst
Benign and malignant tumors Superior vena cava syndrome (arm injection)
Dilated bile ducts Inferior vena cava obstruction (leg injection)
Abscess Focal nodular hyperplasia
Hematoma Budd-Chiari syndrome*
Laceration Cirrhosis (regenerating nodule)*
Localized hepatitis
Radiation therapy *Increased uptake relative to adjacent tissue, not absolute increased uptake.
Infarction
Cirrhosis (pseudotumors)
Fatty infiltration

right lobe, shrinks; the left lobe compensates with

cirrhosis. All may manifest as hepatomegaly; heteroge-
nous uptake and colloid shift ( see Fig. 7-43). The
resulting scan pattern is related to the degree of pathol-
ogy and the presence or absence of portal hyperten-
sion. With increasing severity, the liver, particularly the
hypertrophy and colloid redistribution becomes more
marked (see Fig. 7-43). Splenomegaly with increased
uptake occurs with portal hypertension.
Accuracy
Cold liver lesions can be detected on planar imaging if
they are larger than 2–3 cm and on SPECT if larger than
 Hepatobiliary System 207

Figure 7-45 Superior vena cava syndrome. Left, Hot spot in region of quadrate lobe on Tc-99m
sulfur colloid liver spleen scan in patient with lung cancer and superior vena cava obstruction.
Radiotracer was injected in the arm. Right, Repeat study in same patient with radiotracer injected in
lower extremity. No hot spot is seen.

Figure 7-46 Focal nodular hyperplasia.Anterior (left) and right lateral (right) views show
increased uptake in lesion at inferior tip of right lobe of liver.Angiography confirmed the diagnosis
of focal nodular hyperplasia.

1.5–2 cm. Superficial lesions are more easily detected
than deep ones.
SPECT aids in detecting smaller and more central
lesions because of its improved contrast resolution.
Multiheaded SPECT cameras using ultra-high-resolution
collimators can provide resolution in the range of
1–1.2 cm.
Based on combined data from multiple studies, the
sensitivity for detecting metastatic liver disease with
planar Tc-99m SC imaging is 81% and the specificity is
90%. SPECT improves sensitivity by 10% but retains
specificity.
SPLEEN SCINTIGRAPHY
The spleen serves as a reservoir for formed blood ele-
ments, as a site for clearance of microorganisms and par-
ticle trapping, as a potential site of hematopoiesis during
bone marrow failure, and as a source of humor or cellular
208 NUCLEAR MEDICINE: THE REQUISITES
Figure 7-47 Budd-Chiari syndrome. Relatively increased uptake
of Tc-99m sulfur colloid in region of caudate lobe (arrowheads) in
patient with hepatic vein thrombosis. Images were acquired in the
right lateral (RL), right posterior oblique (RPO), posterior (P), and
anterior (A) projections. Note increased marrow uptake.
response to foreign antigens. It plays a role in leukocyte
production, contributes to platelet processing, and has
immunological functions. Thus the spleen can be visu-
alized by various radiopharmaceuticals with different
mechanisms of uptake, such as Tc-99m SC (RES function),
Figure 7-48 Splenic infarct. Right, Large, wedge-shaped defect (arrowhead) of the spleen in
patient with massive splenomegaly and myeloid metaplasia on Tc-99m sulfur colloid study. Left,
Smaller defects can also be seen on anterior view.
BOX 7-22 Causes of Liver Heterogeneity
of Tc-99m SC Uptake
Metastases (infiltrative, early)
Lymphoma, leukemia
Hepatitis
Fatty metamorphosis
Chronic passive congestion
Parenchymal liver diseases
Cirrhosis
In-111 leukocytes (leukocyte migration),Tc-99m ery-
throcytes (erythrocyte distribution), and damaged RBC
imaging (sequestration).
Radionuclide splenic imaging is used to detect splenic
infarcts (Fig. 7-48), postoperative splenic remnants,
accessory spleens, splenosis, and polysplenia-asplenia
syndromes (Fig. 7-49, A). Although Tc-99m SC scintigra-
phy can often make these diagnoses, liver uptake may
obscure adjacent splenic uptake. In addition, the tip of
the left lobe often migrates into the splenic bed after
splenectomy. Direct CT correlation with the question-
able mass/splenic remnant can aid in diagnosis. Imaging
with heat or chemically damaged Tc-99m erythrocytes
 Hepatobiliary System 209

Figure 7-49 Splenic remnant and splenosis. A, Tc-99m SC with splenic remnant
postsplenectomy best seen in the left lateral view (arrowhead). B, Splenosis, or autotransplantation
of splenic tissue seen after splenic trauma. Damaged Tc-99m-labeled red blood cell study shows
splenic tissue in left upper quadrant. Left to right, Anterior, left lateral, and posterior views.

can provide excellent splenic images with less liver up-
take (Fig. 7-49, B).
Nonvisualization of the spleen may result from con-
genital absence or from acquired functional asplenia
caused by interruption of the blood supply (splenic
artery occlusion) or secondary to RES dysfunction
(sickle cell crisis). Functional asplenia may be irre-
versible (Thorotrast irradiation, chemotherapy, amyloid)
or reversible (sickle cell crisis). With sickle cell disease,
discordance is seen between RES phagocytic function
and other splenic functions.
TC-99M MAA HEPATIC ARTERIAL
PERFUSION SCINTIGRAPHY
Oncologists have used regional intra-arterial chemother-
apy to treat primary and metastatic cancer since the
1960s. Enthusiasm for this form of chemotherapy has
varied over the years, waxing with the introduction of
new technology that makes administration of the
chemotherapy easier, safer, and potentially more effec-
tive and waning after disenchantment with the overall
results in light of the technical difficulties and expense.
210 NUCLEAR MEDICINE: THE REQUISITES

The advantage of a selective intraarterial approach Successful application of intraarterial chemotherapy

to chemotherapy is based on the differential blood
flow to tumor and normal liver. As tumor in the liver
grows, it derives most of its blood supply from the
hepatic artery, whereas normal liver cells are supplied
predominantly by the portal circulation. Intraarterial
chemotherapy delivers the drug preferentially to the
tumor, minimizing exposure to normal liver and to
drug-sensitive, dose-limiting tissues such as gastroin-
testinal epithelium and bone marrow, often the
source of side effects from conventional intravenous
chemotherapy.
requires that the drug be reliably and safely delivered to
the tumor. After initial arteriographic assessment of the
vascular supply of the tumor and liver, a therapeutic
catheter is inserted either: (1) percutaneously, using
a transfemoral or transaxillary approach and attached to
an external infusion pump, or (2) surgically, connected
to a subcutaneously implanted, constant-infusion pump
(Fig. 7-50). Confirmation is needed to ensure that the
perfusion distribution from the catheter truly encom-
passes the entire tumor without perfusion of other vis-
ceral organs (Figs. 7-51 to 7-55).

Figure 7-50 Surgical placement of intraarterial catheters. A, Standard anatomy. Gastroduodenal

artery is ligated and catheter placed at junction of gastroduodenal and common hepatic arteries.
Right gastric artery is ligated. B, Trifurcation. Right and left hepatic arteries originate too close to
gastroduodenal artery to allow equal distribution to all areas of the liver. In this normal variation,
gastroduodenal and right gastric arteries are ligated. Splenic artery is ligated and catheter is
positioned at junction of splenic artery and celiac axis. C, Replaced right hepatic artery originating
from superior mesenteric artery requires use of two catheters. Similarly, patient with left hepatic
artery arising from left gastric artery requires two catheters.
 Hepatobiliary System 211

Figure 7-51 Comparison of Tc-99m macroaggregated albumin (MAA) with Tc-99m sulfur colloid.
Patient with colon cancer and liver metastases. Left, Tc-99m sulfur colloid study shows multiple
lesions involving right and left lobes. Right, Tc-99m MAA study shows solid tumor nodules involving
both lobes of liver in pattern similar to defects seen on Tc-99m SC. Distribution of perfusion is good.
No extrahepatic perfusion.

Figure 7-52 Tc-99m MAA hyperperfusion of peripheral rim of tumor. Left, Tc-99m sulfur colloid
shows large tumor defect in midportion of the liver. Right, Tc-99m MAA study shows hyperperfusion
of the periphery of the large tumor mass with a large, cold, necrotic center.

Figure 7-53 Extrahepatic perfusion. A, Poor perfusion to the left lobe and extrahepatic
perfusion to the stomach. Focal hot spot is caused by partial catheter thrombosis. B, With catheter
replaced, entire liver is well-perfused, although some extrahepatic perfusion to spleen is present.
212 NUCLEAR MEDICINE: THE REQUISITES

Figure 7-54 Extrahepatic perfusion: utility of SPECT. A, Tc-99m sulfur colloid (SC) planar study
shows the left lobe replaced by tumor (cold markers overlie left lobe). B, Tc-99m macroaggregated
albumin (MAA) planar study shows perfusion of the left lobe tumor without definite gastric
perfusion.There is a suggestion of splenic perfusion, and activity adjacent to the left lobe could be
gastric perfusion. C, Tc-99m SC SPECT transverse image shows a large tumor defect in the left lobe.
D, Tc-99m MAA SPECT study shows hyperperfusion of the periphery of the large tumor nodule,
which is cold centrally. Gastric perfusion is seen on the transverse SPECT slice (arrowhead). Splenic
perfusion was seen on other sections not shown here.

Although angiography is needed before initial catheter
placement, it is not a good indicator of blood flow at the
capillary level. The high flow rates required for good con-
trast angiography often do not reflect the actual perfusion
pattern that occurs with the slower infusion rates used in
chemotherapy delivery systems. A high-pressure contrast
bolus may result in streaming, reflux, or retrograde flow.
Contrast angiography cannot be performed through the
small-bore, surgically placed catheters, which deliver
chemotherapy at a rate of 1–5 ml/day.
Incorrect positioning of the intraarterial catheter
results in inadequate perfusion of the tumor-involved
liver and can cause extrahepatic perfusion to the stom-
ach, pancreas, spleen, and bowel (see Fig. 7-53). This can
be caused by difficulties in catheter placement due to
normal vascular anatomical variation. Even if the
catheter is properly placed initially, catheter movement,
occlusion, or arterial thrombosis may produce a change
from the initial perfusion pattern. Tc-99m macroaggre-
gated albumin (MAA) hepatic arterial scintigraphy reli-
ably estimates the adequacy of blood flow to the tumor
and determines the presence or absence of extrahepatic
perfusion, a frequent cause of gastrointestinal and sys-
temic toxicity.
Hepatic arterial perfusion scintigraphy is also used to
quantify lung shunting (prior to using therapeutic Y-90
radiolabeled microspheres [Therasphere and SIR-
Sphere]) to ensure minimal radiation to the lungs.
Mechanism of Localization and
Pharmacokinetics
Tc-99m MAA particles are larger than capillary size
(range, 10–90 μm; mean, 30–50 μm). When infused into

Figure 7-55 Lung uptake of Tc-99m macroaggregated albumin.
Quantitation revealed 40% arteriovenous shunting.
the hepatic artery, they distribute according to blood
flow and are trapped on first pass in the arteriolar-capil-
lary bed of the liver. The irregularly shaped and mal-
leable particles occlude a small percentage of the liver
capillary bed, break down into smaller particles (effec-
tive liver half-life of 4 hours), and are eventually taken up
by macrophages or cleared through the kidney.
Extrahepatic perfusion is seen on Tc-99m MAA perfu-
sion imaging as uptake in abdominal visceral organs,
including the stomach, spleen, and bowel (see Figs. 7-53
and 7-54). Although a small amount of arteriovenous
shunting is common (1–7%),shunting of 10–40% is possi-
ble (see Fig. 7-55). Shunting results in less perfusion of
the tumor, increased systemic exposure, and increased
potential for side effects.
The typical pattern of tumor perfusion on Tc-99m
MAA studies is greater uptake in the tumors compared
with normal liver (tumor:nontumor uptake ratio of 3:1).
Small tumor nodules show uniform uptake (see Fig.
7-51), whereas larger tumors often have increased uptake
at the periphery of the tumor and relatively decreased
uptake centrally because of necrosis (see Fig. 7-52).
Selective hepatic angiography has demonstrated that
most cancers are hypervascular, particularly at the
periphery of the tumor, where active growth occurs
(neovascularity). This increased tumor/nontumor flow
ratio is a major advantage of the intraarterial technique.
Methodology
The method of Tc-99m MAA administration depends on
the type of intraarterial catheter and whether it is placed
percutaneously or surgically (Box 7-23).
Hepatobiliary System 213
Box 7-23 Tc-99m Macroaggregated
Albumin Hepatic Arterial
Scintigraphy: Protocol
Summary
PATIENT PREPARATION
None
INSTRUMENTATION
Camera: large-field-of-view gamma
Collimator: low energy, all purpose, parallel hole
Energy window: 15% centered over 140-keV
photopeak
RADIOPHARMACEUTICAL
Tc-99m MAA, 1–4 mCi (37–148 MBq) for planar imaging
and 5–6 mCi (185–222 MBq) for SPECT
Infuse in small volume (0.5–1 ml) through an
intraarterial catheter
METHOD OF ADMINISTRATION
Surgically Implanted Infusion Pump and Catheter
Insert 22-gauge 1-inch Huber needle into infusion pump
side port.
After ascertaining free flow, infuse Tc-99m MAA slowly
over 1 to 2 min and flush with 10 ml of saline.
Before removing needle, inject 5 ml of heparin
(10 units/ml).
Percutaneously Placed Catheter and External Infusion
Pump
Place three-way stopcock as close as possible to the site
of catheter entry.
With patient positioned under the camera so that
entering flow can be monitored, gently flush catheter
with 10–20 ml of normal saline.
Infuse Tc-99m MAA in 0.2-ml volume via the three-way
stopcock.
Increase the external pump flow rate to 200 ml/hr.
Monitor progress of radioactive injectant on the
persistence scope.As bolus approaches the liver,
decrease flow rate of the pump to rate at which
chemotherapy is to be delivered, generally 10–20
ml/hr.
IMAGING PROTOCOL
Acquire images with the patient lying on the table
supine.
Acquire 500k-count anterior image, then posterior, right
and left lateral, and anterior chest views for equal
time.
If extrahepatic gastric perfusion is suspected, 4 g of
sodium bicarbonate–citric acid–simethicone
effervescent granules should be given in 100 ml of
water by mouth.The patient must be encouraged not
to belch. Repeat anterior and left lateral images.
SPECT option: technique similar to Tc-99m sulfur colloid
SPECT.
214 NUCLEAR MEDICINE: THE REQUISITES
Clinical Applications
The Tc-99m MAA hepatic arterial perfusion study is often
performed after initial catheter placement and before
subsequent courses of chemotherapy, particularly if the
patient has symptoms suggestive of gastrointestinal tox-
icity. Effectiveness of intraarterial chemotherapy is maxi-
mized if the entire tumor-involved liver is perfused and
side effects are minimized if there is no extrahepatic per-
fusion or AV shunting to the lung.
Symptoms (e.g., pain, nausea, vomiting) caused by
tumor involvement may be difficult to differentiate
from those caused by extrahepatic perfusion of the
stomach and bowel. The importance of extrahepatic
perfusion is the high associated incidence of adverse
symptoms (70% vs. 20% in patients without extrahepatic
perfusion), including nausea, vomiting, gastritis, ulcera-
tion, and hemorrhage.
Image Interpretation
Hepatic uptake is typically inhomogeneous. Tumor nodules
have increased uptake compared with surrounding normal
liver. Multiple planar views (right lateral, anterior, posterior,
left lateral) or SPECT can determine the perfusion distribu-
tion. Comparison with a recent CT can be helpful.
Arteriovenous shunting is noted as lung uptake (see
Fig. 7-55). Tc-99m-MAA particles shunted through the
tumor bypass the liver capillary bed and are trapped in
the lung. Lung shunting gives an estimate of the percent-
age of drug not delivered to the tumor that may result in
systemic exposure and toxicity.
SUGGESTED READING
Choy D, Shi EC, McLean RG, et al: Cholescintigraphy in
acute cholecystitis: use of intravenous morphine. Radiology
151: 203-207, 1984.
Fig LM, Stewart RE,Wahl RL: Morphine-augmented hepatobil-
iary scintigraphy in the severely ill: caution is in order.
Radiology 175: 473-476, 1990.
Freitas JE, Coleman RE, Nagle CE, et al: Influence of scan and
pathologic criteria on the specificity of cholescintigraphy.
J Nucl Med 24: 867-879, 1983.
Sostre S, Kaloo AN, Spiegler EJ, et al: A noninvasive test of
sphincter of Oddi dysfunction in postcholecystectomy
patients: the scintigraphic score. J Nucl Med 33: 1216-1222,
1992.
Weissmann HS, Freeman LM: The biliary tract. In Freeman and
Johnson’s clinical radionuclide imaging. Freeman LM, Ed.
New York, Grune & Stratton, 1984.
Yap L,Wycherley AG,Morphett AD,Toouli J: Acalculous biliary pain:
cholecystectomy alleviates symptoms in patients with abnormal
cholescintigraphy. Gastroenterology 101:786-793,1991.
Zeman RK,Lee C,Stahl RS,et al: Ultrasonography and hepatobil-
iary scintigraphy in the assessment of biliary-enteric anasto-
moses. Radiology 145: 109-115, 1982.
Zeman RK, Lee C, Jaffe MH, Burrell MI: Hepatobiliary scintigra-
phy and sonography in early biliary obstruction. Radiology
153: 793-798, 1984.
Ziessman HA, Zeman RK, Akin EA: Cholescintigraphy:
Correlation with other hepatobiliary imaging modalities. In
Diagnostic Nuclear Medicine, 4th ed. Philadelphia, Lippincott
Williams & Wilkins, 2003.
Ziessman HA, Silverman PM, Patterson J, et al: Improved detec-
tion of small cavernous hemangiomas of the liver with high-
resolution three-headed SPECT. J Nucl Med 32:2086-2091,1991.
Ziessman HA, Fahey FN, Hixson DJ: Calculation of a gallblad-
der ejection fraction: advantage of continuous sincalide infu-
sion over the 3-minute method. J Nucl Med 33: 537-541,
1992.
Ziessman HA, Muenz LR,Agarwal AK, ZaZa A: Normal values for
sincalide cholescintigraphy: comparison of two methods.
Radiology; 221: 404-410, 2001.
Ziessman HA,Thrall JH,Yang PJ, et al: Hepatic arterial perfusion
scintigraphy with Tc-99m MAA. Radiology 152: 167-172, 1984.
Ziessman HA: Acute cholecystitis, biliary obstruction, and bil-
iary leakage. Sem Nucl Med 33:279-296, 2003.
Ziessman HA, Jones DA, Muenz LR,Agarval AK: Cholecystokinin
cholescintigraphy: methodology and normal values using a lac-
tose-free fatty-meal food supplement. J Nucl Med 33: 1263-
1266, 2003.
 8
CHAPTER Genitourinary System

Renal Anatomy and Function Methodology

Renal Radiopharmaceuticals Dosimetry
Iodine-131 Orthoiodohippurate Image Interpretation
Chemistry and Radiolabeling Scrotal Scintigraphy
Pharmacokinetics Radiopharmaceutical
Special Considerations Methodology
Tc-99m Diethylenetriamine Pentaacetic Acid Image Interpretation
Chemistry and Radiolabeling Normal Pattern
Pharmacokinetics Acute Testibular Torsion
Tc-99m Mercaptoacetyltriglycine Delayed Torsion
Radiolabeling Acute Epididymitis
Pharmacokinetics Torsion of the Testicular Appendage
Tc-99m Dimercaptosuccinic Acid and Tc-99m Glucoheptonate
Radiolabeling Early renal radionuclide studies in the 1950s did not
Pharmacokinetics image the kidneys but used external gamma probes
Dosimetry to generate time-activity clearance histograms. Probe
Dynamic Renal Imaging Techniques studies subsequently gave way to gamma camera imaging.
Dynamic Renography Because of advancements in magnetic resonance imaging
Methodology (MRI), computed tomography (CT ), and ultrasound,
Computer Processing nuclear medicine techniques are mostly utilized for func-
Interpretation tional renal imaging and quantification, often providing
Clinical Applications of Renal Scintigraphy information not possible with the anatomic or structural
Renovascular Hypertension modalities. Some examples include: evaluation of blood
Urinary Tract Obstruction flow and viability, differentiation of obstructive and
Renal Transplant Evaluation nonobstructive hydronephrosis,confirming urinary leaks,
Measuring Renal Function: ERPF and GFR and diagnosing acute pyelonephritis.
Effective Renal Plasma Flow Other indications for renal scintigraphy (Box 8-1)
Glomerular Filtration Rate include evaluation of viability, infection, and masses diffi-
Camera Technique cult to assess by other methods such as ultrasound. The
Renal Cortical Imaging ability to quantify function by effective renal plasma flow
Acute Pyelonephritis and glomerular filtration rate can provide a better meas-
Methodology ure of functional status than standard laboratory tests
Image Interpretation such as creatinine clearance. For each different diagnos-
Clinical Applications tic problem, different radiopharmaceuticals have been
Radionuclide Cystography developed ( Table 8-1). This chapter concentrates on
Vesicoureteral Reflux agents that have been used clinically.

215
216 NUCLEAR MEDICINE: THE REQUISITES

rounds the glomerulus. Each kidney contains more than

Box 8-1 Clinical Indications for 1 million of these basic functional units of the kidney, the
Genitourinary Scintigraphy nephron.
Normally, the kidneys receive 20% of cardiac output
Perfusion abnormalities with renal plasma flow ( RPF ) averaging 600 ml/min.
Acute and chronic renal failure The kidneys clear the plasma and body of waste prod-
Renal transplant: rejection, obstruction, status of ucts. The clearance, or rate of disappearance, of a sub-
anastomosis stance can be measured as:
Renal trauma or surgical complications
Renovascular hypertension/renal artery stenosis urine concentration (mg /ml) ×
Quantification of renal function: GFR/ERPF urine flow (ml /min)
Pyelonephritis Clearance (ml/min) =
Mass vs. column of Bertin plasma concentration (mg/ml)
Ureteral obstruction
Vesicoureteral reflux Plasma clearance occurs by glomerular filtration and tubu-
Bladder residual volume quantification lar secretion (Fig. 8-2). If an agent undergoes 100% first
pass extraction, then it can be used to measure RPF.
However, because extraction of agents used to measure
renal function are less than 100%, the term effective renal
plasma flow ( ERPF ) is used to describe the measurement.
Table 8-1 Historical Perspective on Radionuclides Approximately 20% of RPF (120 ml/min) is filtered
for Renal Function Evaluation through the semipermeable membrane of the glomeru-
lus. A pressure gradient created by the RPF and resis-
tance in the vessel drives filtration from the vascular
Year Radiopharmaceutical Method
space into the renal tubules. Larger material such as pro-
1952 I-131 Iopax Urine counting tein-bound compounds will not be filtered, whereas
1956 I-131 Diodrast Renogram small molecules with a polar charge will be filtered. The
1960 I-131 hippuran Renogram resulting ultrafiltrate consists of water and crystalloids
1968 I-131 hippuran Lasix renography
1968 Hg-203 chloride Individual renal function
but no colloids or cells. The molecule inulin is considered
1969 Tc-99m gluconate Renal scan the gold standard for glomerular filtration measurement.
1970 Tc-99m DTPA Renal scan, GFR The remaining plasma moves into the efferent arteri-
1971 I-131 hippuran Single-sample ERPF ole where active secretion occurs at the tubular epithe-
1974 Tc-99m DMSA Renal scan lial cells. Molecules that could not be filtered may be
1984 I-131 hippuran Captopril renography
1986 Tc-99m MAG3 Renal scan
cleared into the tubular lumen by active tubular secre-
tion. Overall, tubular secretion accounts for 80% of renal
plasma clearance. Paraaminohippurate ( PAH) is the clas-
sic method for measuring ERPF because its high extrac-
tion mirrors the distribution of RPF: 20% of PAH is
RENAL ANATOMY AND FUNCTION cleared by glomerular filtration and 80% is secreted into
the renal tubules. PAH is actively secreted by anionic
The kidneys are responsible for regulating water and transporters on the proximal convoluted tubular cell
electrolyte balance, excreting waste, secreting hormones membranes. It is not metabolized or retained in the kid-
(renin, erythropoietin), and activating vitamin D. They lie ney and is not highly protein-bound,so plasma extraction
in the posterior abdomen at the level of the first to third is high. However, PAH does have some plasma protein
lumber vertebra. The right kidney often lies slightly infe- binding, and clearance is roughly 85–95%.
rior to the left. The outer cortex contains the glomeruli When urine passes along the tubule, essential sub-
and proximal convoluted tubules. The renal pyramids stances such as glucose,amino acids,and sodium are con-
consisting of the collecting tubules and the loops of served. The filtrate is concentrated as 65% of the water
Henle make up the medulla. At the apex of the pyramids, filtered at the glomerulus undergoes reabsorption in the
papillae drain into the renal calyces. Cortical tissues proximal convoluted tubule. Active sodium pumping in
between the pyramids are known as the columns of the loop of Henle sets up an osmotic gradient allowing
Bertin ( Fig. 8-1, A). water to passively diffuse back into the interstitium. The
The renal artery supplies the blood flow to the kidney. remaining concentrated urine passes down the renal
End arterioles lead to tufts of capillaries forming tubule, through the papillae of the medullary pyramids
glomeruli in the renal cortex ( Fig. 8-1, B). The most prox- into the calyces. The calyces empty into the renal pelvis,
imal end of the renal tubule, Bowman’s capsule, sur- and urine passes down the ureter into the bladder.
 Genitourinary System 217

Columns
of Bertin Distal
Pyramid Interlobar Efferent convoluted
artery arteriole tubule
Afferent
arteriole
Papilla Arcuate
Glomerulus
arteries
Bowman's capsule
Calyx Straight
arteries Proximal
(interlobular) convoluted
tubule

Renal Glomeruli
artery
Renal
vein
Collecting
tubule
Pelvis
Arcuate artery Descending
and vein and ascending
loops of
Henle

Ureter

A B
Figure 8-1 Renal anatomy. A, The outer layer or cortex is made up of glomeruli and proximal
collecting tubules.The inner layer, or medulla, contains pyramids, made up of distal tubules and
loops of Henle.The tubules converge at the papillae, which empty into calyces.The columns of
Bertin, between the pyramids, are also cortical tissue.The renal artery and vein enter and leave at the
hilus.The interlobar branches of the renal artery divide and become the arcuate arteries, which give
rise to the straight arteries, from which arise the afferent arterioles that feed the glomerular tuft.
B, The nephron consists of afferent vessels leading to the tuft of capillaries in the glomerulus, the
glomerulus itself, and efferent vessels. Bowman’s capsule surrounds the glomerulus and connects to
the proximal and distal renal tubules and loops of Henle.

other renal radiopharmaceuticals (Fig. 8-4). As a radio-

RENAL RADIOPHARMACEUTICALS
Renal radiopharmaceuticals are classified by their uptake
and clearance mechanisms as agents for glomerular filtra-
tion, tubular secretion, or cortical binding ( Table 8-2). The
first step in image interpretation is understanding uptake
physiology ( Fig. 8-3). Renal radiopharmaceuticals are
divided into those measuring glomerular filtration
(Box 8-2),tubular secretion ( Box 8-3),and cortical binding.
The most important renal agents clinically are Tc-99m
diethylenetriamine pentaacetic acid ( Tc-99m DTPA), I-131
hippuran ( I-131 OIH ), Tc-99m mercaptylacetyltriglycine
(Tc-99m MAG3),Tc-99m glucoheptonate ( Tc-99m GH ),
and Tc-99m dimercaptosuccinic acid ( Tc-99m DMSA).
Iodine-131 Orthoiodohippurate
I-131 OIH has long been a valuable tool for evaluating
renal function. Although it has now been replaced in
clinical use, it remains the foundation for discussion of
pharmaceutical chemically and pharmacokinetically sim-
ilar to PAH, I-131 OIH is cleared by tubular secretion
(80%) and by glomerular filtration (20%). The first pass
extraction is high, roughly 85%. Therefore, hippuran
accurately measures ERPF. The high level of clearance
made I-131 OIH useful for imaging azotemic patients.
Chemistry and Radiolabeling
Hippuran can be radiolabeled with either I-131 or I-123
through an exchange reaction. Stabilizing and buffering
agents are added for pH adjustments. The bond tends to
degrade with time. Storage should be at 4°C or less, pro-
tected from light.
Pharmacokinetics
With normal renal function, peak renal concentration is
reached within 2–4 minutes of intravenous injection. Sub-
sequent visualization in the collecting system and bladder
is faster than with Tc-99m DTPA. The time it takes for the
activity to be cleared can be described in terms of the time
218 NUCLEAR MEDICINE: THE REQUISITES

100%
100%

Afferent
arteriole

Proximal
tubule

Urine
20% 20% Urine
Glomerulus
20% excreted 100%
80% excreted

80%
80% Tubular
cell

Efferent
arteriole
B
A
Figure 8-2 Renal plasma flow and function. A, Glomerular filtration.Twenty percent of renal
blood flow to the kidney is filtered through the glomerulus. B, Tubular secretion.The remaining 80%
of renal plasma flow is secreted into the proximal tubules from the peritubular space.

Glomerular filtration: Tubular secretion:

51Cr-EDTA,99mTc-DTPA, 123I-OIH,131I-OIH,
Table 8-2 Mechanism of Uptake for Renal 99mTc-MAG3
125I-iothalamate
Scintigraphy Agents

Uptake Agent

Glomerular filtration (100%) Tc-99m DTPA

Tubular (100%) Tc-99m MAG3
Tubular (80%) and glomerular I-131 and I-123
(20%) hippuran
Cortical binding (40%) Tc-99m DMSA
Glomerular filtration (80%) and Tc-99m
cortical binding (20%) Glucoheptonate
Tubular fixation:
99mTc-DMSA,
99mTc-glucoheptonate

required to clear 50% of the activity after reaching the peak Cortex
cortical activity. This clearance half-time, or T ⁄ , for I-131
1
Medulla
2

OIH is normally 10–15 minutes with 98% of the dose

cleared from plasma within 24 hours.

Special Considerations
The long half life (8.08 days) and beta-radiation of
the I-131 label necessitate a low administered dose
of 200–400 uCi (7.4–14.8 MBq). This dose limit and
the high-energy gamma photon (364 keV ) contribute to
poor spatial resolution and an inability to do dynamic
Figure 8-3 Different mechanisms of renal radiopharmaceutical
arterial perfusion imaging. The dose to the thyroid from uptake and excretion.These include glomerular filtration, tubular
unlabeled radioiodine can be reduced by pretreatment secretion, and cortical tubular binding.

Box 8-2 Agents Used for Glomerular
Filtration
C-14 or H-3 inulin
I-125 diatrizoate
I-125 iothalamate
Co-57 vitamin B12
Cr-51 EDTA
In-111 DTPA
Yb-169 DTPA
Tc-99m DTPA
Box 8-3 Renal Tubular
Radiopharmaceuticals Used to
Quantify Effective Renal Plasma
Flow
H-2 or C-14 paraaminohippurate (PAH)
I-125, I-131 iodopyracet
I-125, I-123, or I-131 orthoiodohippurate (OIH)
Tc-99m mercaptoacetyltriglycine (MAG3)
with thyroid blocking agents like supersaturated potas-
sium iodide (SSKI ). The United States Pharmacopeia
(USP ) mandates there be less than 3% free radioiodine. It
was thought that the limitations created by the I-131 label
would be overcome by utilizing I-123. Although theoreti-
cally a better label, there were logistical problems con-
cerning timely distribution of I-123. Also, surprisingly
high dosimetry limited the administered dose to 3 mCi,
which resulted in poor flow studies. However, I-131 OIH
has been replaced by technetium-99m labeled MAG3 and
I-131 OIH is no longer available commercially.
Tc-99m Diethylenetriamine Pentaacetic Acid
The physical characteristics of the technetium label of
Tc-99m DTPA allow dynamic arterial perfusion studies
and much better discrimination of cortex and collecting
system than I-131 OIH.
Chemistry and Radiolabeling
DTPA is a heavy metal chelator used for treatment of poi-
soning. Like other chelators, it is cleared through
glomerular filtration. Tc-99m DTPA is convenient and
inexpensive to prepare. A simple kit is available contain-
ing stannous tin as a reducing agent. The reduced tech-
netium-99m forms a powerful chelating bond with the
DTPA. Contaminants such as hydrolyzed and oxidized
Genitourinary System 219
technetium-99m (free pertechnetate) can be readily iden-
tified by chromatography.
Pharmacokinetics
Following intravenous injection of Tc-99m DTPA, normal
peak cortical uptake occurs by 3–4 minutes. By 5 min-
utes, the collecting system is seen; the bladder is typically
visualized by 10–15 minutes. The T ⁄ peak, or the time it
1
2
takes for one-half of the maximal cortical activity to clear,
is normally 15–20 minutes for Tc-99m DTPA.
Tc-99m DTPA is essentially completely filtered at
the glomerulus with no tubular secretion or reabsorption.
Because only 20% of renal function is the result of glomeru-
lar filtration, the first-pass extraction of a glomerular filtra-
tion agent is less than agents cleared by tubular secretion. In
practice, normal first-pass Tc-99m DTPA extraction is less
than the 20% of RPF due to factors such as the level of pro-
tein binding. The extraction fraction is only 40–50% that of
Tc-99m MAG3 and is even lower compared to I-131 OIH.
The lower extraction efficiency becomes clinically signifi-
cant in patients with abnormal renal function or obstruc-
tion. In such cases, target-to-background ratios may be so
poor that no diagnostic information is gained. At many sites,
Tc-99m DTPA use is reserved for glomerular filtration rate
(GFR) calculations, whereas blood flow and routine func-
tional assessments are done with Tc-99m MAG3.
The rate of Tc-99m DTPA clearance depends on the
amount of impurities in the preparation which bind to
protein in the body. This leads to underestimation of
GFR. Different preparations of Tc-99m DTPA may show
highly variable protein binding that should be con-
trolled. Although Tc-99m DTPA generally underestimates
GFR, it is adequate for clinical use if the level of protein
binding is minimized.
Tc-99m Mercaptoacetyltriglycine
Currently,Tc-99m MAG3 is the most commonly used renal
radiopharmaceutical. It is cleared almost entirely by tubu-
lar secretion. The extraction efficiency is considerably
higher than filtration agents. This results in better perform-
ance when function is compromised. The improved count-
ing statistics of Tc-99m MAG3 over I-131 OIH permit blood
flow imaging. In addition, the high-count time–activity
curves and images are far superior with lower radiation
dosimetry (Fig. 8-5). Tc-99m MAG3 images show signifi-
cant anatomic detail while assessing function (Fig.8-6).
Radiolabeling
Tc-99m MAG3 is available in kit form. Labeling involves
the addition of sodium pertechnetate to a reaction vial.
A unique feature of the Tc-99m MAG3 labeling process
is that a small amount of air is added to the reaction vial
to consume excess stannous ion for increased stability.
Radiolabeling efficiency is 95% or greater.
220 NUCLEAR MEDICINE: THE REQUISITES

Figure 8-4 Radiopharmaceutical comparison in a renal transplant patient. A, I-131 hippuran

provides excellent functional information but has poor image quality compared to technetium
agents. B, Tc-99m DTPA image from the same day shows higher resolution. C, Tc-99m MAG3 done
30 hours later reveals the highest level of detail as well as an improved target to background ratio
compared to DTPA.
 Genitourinary System 221

Renogram
Cts
19976

14982
Left kidney
Right kidney
Bladder
9988

4994

0
0 7 14 20 27
Time (min)
B
Figure 8-5 Normal Tc-99m MAG3. A, Normal dynamic functional images with prompt symmetric
radiotracer uptake and rapid clearance over the study. B, Normal time–activity curves with a steep
uptake slope, distinct peak, and rapid clearance confirming image analysis.
222 NUCLEAR MEDICINE: THE REQUISITES

Pharmacokinetics
Because Tc-99m MAG3 is protein-bound and not filtered, it
is exclusively cleared from the kidney by tubular secretion.
It has a much higher first-pass extraction than a glomerular
filtration agent like Tc-99m DTPA. Plasma protein binding is
97% for Tc-99m MAG3 as opposed to 70% for I-131 OIH.
This binding keeps Tc-99m MAG3 in the vascular space,
improving renal target-to-background ratios but slowing
renal extraction. The clearance is only about 60% of I-131
OIH. The alternative route of Tc-99m MAG3 excretion is via
the hepatobiliary route. Liver activity and biliary tract clear-
ance are frequently noted. The normal time to peak activity
is 3–5 minutes. Normal clearance and half-peak values are
similar to other agents described.
Cortical Binding Radiopharmaceuticals:
Technetium-99m Dimercaptosuccinic Acid
and Technetium-99m Glucoheptonate
Renal cortical imaging evaluates viability, infection, and
structural abnormalities difficult to assess by ultrasound
and CT. The original cortical imaging agents used
a diuretic, chlormerodrin, labeled with mercury-203 and
mercury-197. These were abandoned in favor of Tc-99m
DMSA (a chelating agent) and Tc-99m GH (a former con-
ventional brain scan agent). Tc-99m DMSA is the better
agent for evaluating the renal cortex (Figs.8-7 and 8-8).
The rapid transit of most renal radiopharmaceuticals
(Tc-99m DTPA, I-131 OIH, and Tc-99m MAG3) does not

Figure 8-6 Abnormal Tc-99m MAG3 examples revealing anatomic and functional data.A, Small
scarred left kidney secondary to vesicoureteral reflux contributing 15% to overall function.Good
cortical clearance is seen.B, Obstruction of right kidney secondary to cervical carcinoma. A neph-
rostomy tube is draining well and bilateral function is good.Prominent left pelvis and calyces mostly
clear by the end of the study;this study shows hydronephrosis but no significant obstruction.The last
image was taken with the bladder in view.
Continued
 Genitourinary System 223

Figure 8-6, cont’d C, Ureteral leak (arrow) detected on sequential images in a

postoperative patient. D, Duplicated right collecting system, a congenital abnormality
associated with lower pole reflux and upper pole obstruction.
224 NUCLEAR MEDICINE: THE REQUISITES

Figure 8-7 Tc-99m DMSA planar and SPECT studies in two patients with cortical scars caused by
reflux. A, Planar images were acquired using a pinhole collimator. Left to right, Left posterior
oblique (LPO), posterior (left kidney), posterior (right kidney), and right posterior oblique views.
Cortical defect in the left superior pole (arrowhead) is best seen on the LPO view. B, High-
resolution SPECT. Sequential 3.5-mm coronal sections show a cortical defect in the right upper pole
and a larger defect in the right lower pole (arrowheads). Note a distinct separation of cortex from
medulla and collecting system with SPECT.

allow high-resolution imaging of the cortex. On the
other hand, the stable cortical uptake of Tc-99m DMSA
and Tc-99m GH produces high-quality images using
either pinhole imaging or single-photon emission com-
puted tomography (SPECT ). Delayed imaging results in
high target-to-background ratios and good resolutions.
Radiolabeling
Both agents are available for preparation in a kit form
using Tc-99m pertechnetate reduced by stannous ion.
The introduction of air can lead to degradation of the
label and lead to increased background activity, including
in the liver. Tc-99m DMSA radiolabeling produces multi-
ple forms of the Tc-99m DMSA complex which may vary
slightly in their clearance.
Tc-99m GH Pharmacokinetics
Approximately 80% of the injected dose of Tc-99m GH is
cleared into the urine, with approximately 15–20% of the
dose fixed in the proximal cortical tubules. Imaging is done
in two stages. Dynamic assessment of flow and function is
obtained in the first 25–30 minutes after injection. An imag-
ing sequence similar to Tc-99m DTPA or Tc-99m MAG3 is
used. Following dynamic imaging,delayed static images are
performed. Renal uptake depends on RPF and renal tubular
function. An alternative route of excretion may occur
through the liver and gallbladder filling may occur.
Tc-99m DMSA Pharmacokinetics
The mechanism of Tc-99m DMSA renal uptake and clear-
ance is not entirely understood. Roughly 40% of the
injected dose localizes in the cortex, predominantly in the
proximal tubules. Imaging is generally done after a 2- to
3-hour delay to allow time for uptake and slow background
clearance. In cases of decreased renal function, further
delay may be needed. The low level of urinary excretion
(approximately 25% of the dose) is not adequate for assess-
ment of the collecting system and lower urinary tract.
Diseases affecting the proximal tubules, such as renal
tubular acidosis and Fanconi’s syndrome, inhibit Tc-99m
 Genitourinary System 225

Figure 8-8 Tc-99m glucoheptonate imaging in a patient with vesicoureteral reflux, posterior
views. A, Dynamic planar study. B, Delayed cortical imaging. Both phases show a cortical defect in the
left lower pole. Delayed images acquired with a pinhole collimator which improved resolution. Left to
right: Right posterior oblique, posterior, and left posterior oblique views.
226 NUCLEAR MEDICINE: THE REQUISITES

DMSA uptake. Nephrotoxic drugs including gentamicin Method

and cis-platinum also inhibit uptake. In cases where
serum creatinine is significantly elevated, target-to-back-
ground ratios may be so poor that no useful diagnostic
information can be gained.
Tc-99m DMSA is preferred over Tc-99m GH because
its greater cortical uptake results in superior image reso-
lution. Therefore, Tc-99m GH is currently rarely used.
However,Tc-99m GH may be useful in patients with renal
tubular acidosis because Tc-99m GH uptake is not
affected, unlike with Tc-99m DMSA.
Dosimetry
Although the radiation dose to the patient from renal radio-
pharmaceuticals is low when renal function is normal, the
absorbed dose rises significantly in the obstructed kidney
or when renal function fails. Dosimetry of the important
renal radiopharmaceuticals is listed in Table 8-3.
DYNAMIC RENAL IMAGING
TECHNIQUES
Renal protocols must be tailored for specific clinical
applications. This section provides a basic approach
to these techniques and modifications. The protocols
outline current use of the major clinical radiopharma-
ceuticals.
Dynamic Renography
Dynamic functional studies are generally acquired in two
parts. The renal blood flow is assessed in the first pass of
the radiopharmaceutical bolus to the kidney. Then over
the following 25–30 minutes, uptake and clearance
assess function.
Similar protocols can be used for the dynamic func-
tional agents Tc-99m DTPA,Tc-99m GH,and Tc-99m MAG3
( Box 8-4). I-131 OIH count rates are insufficient for blood
flow studies as the amount of administered activity is lim-
ited due to the higher absorbed radiation doses.
Patient Preparation
Patients should be well-hydrated before the study.
Dehydration does not affect blood flow, radiopharmaceu-
tical uptake, or functional calculations (ERPF or GFR).
However,excretion and washout can be delayed by dehy-
dration, simulating obstruction.
It is important to document all medications the
patient has taken that may affect the study, such as
diuretics and blood pressure medicines. Any known
anatomic anomalies and prior interventions are impor-
tant factors to consider in positioning and image inter-
pretation.
Patient Positioning
A supine position is preferred as kidneys are frequently
mobile (or “ptotic”) and can move to the anterior pelvis
when patients are upright. Patients are placed so that the
kidneys are closest to the camera, with the camera poste-
rior for normal native kidneys and anterior for trans-
plants, pelvic kidneys and horseshoe kidneys.
Dose
The dose of radiopharmaceutical varies with the agent.
Typical doses are listed in the protocol boxes. In the
past, higher doses of Tc-99m MAG3 up to 10–15 mCi
were given, but as little as 3–6 mCi (111–222 MBq) is
actually needed in adults. Although nomograms are avail-
able for calculating pediatric doses based on body sur-
face area,Webster’s rule is generally helpful: [age + 1]/
[age + 7] × adult dose.
Image Acquisition
Following a bolus injection of radiopharmaceutical, the
image acquisition begins when activity is about to enter
the abdominal aorta. Images are acquired at a rate of 1–3
seconds per frame for 60 seconds to assess renal perfu-
sion. Then images are obtained at 60 seconds per frame
for 25–30 minutes to evaluate parenchymal radiotracer
uptake and clearance.
Computer Processing of Renal Studies
The uptake and clearance of radiopharmaceuticals
is a dynamic process. Mentally integrating all the infor-
mation in the many images of a renal scan is challeng-

Table 8-3 Radiation Dosimetry for Renal Radiopharmaceuticals in rad/mCi (cGy/MBq)

Organ I-123 OIH I-131 OIH Tc-99m DTPA Tc-99m MAG3 Tc-99m DMSA Tc-99m GH

Bladder 0.95 5.167 0.27 0.6 0.07 0.28

Kidneys 0.05 0.167 0.09 0.0175 0.63 0.17
Ovaries 0.05 0.1 0.016 0.0325 0.013 0.016
Testes 0.03 0.067 0.01 0.02 0.007 0.01
Whole body 0.02 0.03 0.006 0.01 0.015 0.0075

Box 8-4 Protocol Summary for Dynamic
Renal Scintigraphy
PATIENT PREPARATION
Hydration
Adults: drink 300–500 ml of water
Children: Intravenous hydration 10–15 ml/kg over 30
min; <1 year use dextrose 5% in 0.45% normal saline,
>1 year of age D5 in 0.45% normal saline
Patient must void before starting study
RADIOPHARMACEUTICAL
Tc-99m DTPA
Adults: 15 mCi (555 MBq)
Children: 200 UCi/kg (2 mCi minimum, 10mCi
maximum)
Tc-99m GH
Adults: 20 mCi (740 MBq)
Children: same as for Tc-99m DTPA
Tc-99m MAG3
Adults: 3–5 mCi (110–185 MBq)
Children: 100 UCi/kg (1 mCi minimum to 5 mCi
maximum)
INSTRUMENTATION
Camera: large field of view gamma
Collimator: low energy, parallel hole
Photopeak: 15–20% window centered over 140 keV
PATIENT POSITION
Routine renal imaging: supine, posterior
Renal transplant: patient supine, camera anterior over
allograft
COMPUTER ACQUISITION
Blood flow: 1- to 2-sec frames for 60 sec
Dynamic: 30-sec frames for 25 min
Pre-void image 500k count
Postvoid image
PROCESSING
Draw computer region of interest around kidneys and
for background area
Generate time–activity curves for 60-sec flow phase and
for 25-min dynamic study
ing, even for experienced clinicians. Computer-
generated time–activity curves ( TAC) provide a dyna-
mic visual presentation of changes in activity over the
course of the study. Usually separate TACs are drawn
for the blood flow and dynamic function portions of
the study.
Perfusion Time–Activity Curve
The first-pass perfusion TAC shows the blood flow to
each kidney compared with arterial flow. A region of
Genitourinary System 227
interest is drawn around each kidney and the closest
major artery (aorta for native kidneys, iliac artery for
transplanted kidneys) on the initial 60-second portion of
the study. Although absolute flow (measured in
milliliters per kilogram per minute) cannot be calculated
with the radiotracers discussed, relative flow can be visu-
alized or calculated using the upslope of the perfusion
curve. A ratio of the activity compared to the aorta or
K/A ratio can help follow changes in perfusion.
Dynamic Renal Function Time–Activity Curve
Evaluation of dynamic cortical function is done with
a renal region of interest ( ROI ) corrected for background
activity ( Fig. 8-9). The selection of kidney ROI depends
on the information needed. Whole kidney ROIs can be
used for cortical function if the collecting system clears
promptly. When a whole kidney ROI is used in a patient
with retained activity in the collecting system, clearance
will appear delayed on the time–activity curve ( TAC). In
cases of hydronephrosis and obstruction, a 2-pixel wide
peripheral cortical ROI excluding collecting system is
best, although some overlap with calyces is inevitable.
Various methods of background correction have been
employed. A 2-pixel wide region of interest is drawn. It
may be placed beneath the kidneys, around the kidneys,
or in a crescent configuration. Because the kidneys over-
lap the liver and spleen, most background correction
methods include some liver and spleen in the back-
ground ROI. Background correction is less critical in
delayed images, such as those obtained with Tc-99m
DMSA, because of the high target-to-background ratio on
these delayed images.
At any point in time, the renogram represents a sum-
mation of uptake and excretion. Three phases are nor-
mally seen in the TACs. These include blood flow,
cortical uptake, and clearance phases ( Fig. 8-10). The
TAC must be interpreted in conjunction with the
images as the curves may be affected by many factors,
such as retained activity in hydronephrosis, which can
alter the slope. Any technical error or discrepancy
between the appearance of the curve and the images
must be reconciled.
Semiquantitative Assessment of Renal Function
Numerous values can be derived from the time–activity
curves. These include time to peak activity, uptake slope,
rate of clearance, and percent clearance at 20 minutes.
The most commonly used parameters reflecting clear-
ance are the 20/peak ratio and the 20/3 ratio. The
20/peak is the ratio of cortical activity at 20 minutes to
the amount of peak activity. The 20/3 ratio is calculated
by dividing the amount of activity at 20 minutes by the
activity at 3 minutes.
Differential Function
Differential or split function is a universally performed
calculation. A whole kidney ROI is needed so that no cor-
tex is excluded. The ROI is related to generate a TAC
228 NUCLEAR MEDICINE: THE REQUISITES

Figure 8-9 Regions of interest (ROI) for time–activity curves. Left, An image at 3 minutes with
peak cortical activity is chosen for the ROIs. Right, Regions of interest are drawn for the kidney
(dark lines) and for background correction (gray lines).

2
Counts
Counts

L
3
1

0 8 16 24
Time (min) Time
Figure 8-10 Normal renogram can be divided into phases. 1, Initial blood flow (30–60 seconds).
2, Cortical uptake phase (normally 1–3 minutes). 3, Clearance phase representing cortical excretion
and collecting stem clearance.

representing the amount of activity in each kidney dur- Interpretation

ing the peak nephrogram phase. Values are selected in
the range of 1–3 minutes after injection (after the blood
flow phase and before the excretory phase). The actual
counts in each kidney are expressed as a fraction of 100%
total function. Normally, the relative contribution for
each kidney lies between 45–55%. This value does not
indicate if the overall or global renal function is normal
or abnormal. A calculation of GFR or ERPF can be done
as a separate study to quantify actual function.
Flow Phase
Blood flow to the kidneys is normally seen within 2–5 sec-
onds of abdominal aorta visualization. It is important to
assess the quality of the injection bolus, as delayed renal
visualization may be artifactual due to suboptimal injection
technique. If the slope of the arterial time–activity curve is
not steep or if activity visibly persists in the heart and
lungs, the injection may have been too strung out. Any
asymmetry in tracer activity suggests abnormal perfusion
 Genitourinary System 229

to the side of decreased or delayed activity. A smaller or
scarred kidney will have less flow due to a decrease in
parenchymal tissue volume ( Fig. 8-11).
Cortical Function Phase
Like the nephrogram phase of an intravenous pyelogram
(IVP), normal kidneys accumulate radiopharmaceutical
in the parenchymal tissues in the first 1–3 minutes. The
cortex should appear homogeneous. The calyces and
renal pelvis are either not seen in this initial phase or
appear relatively photopenic. If decreased function is
present on one side, the rate of uptake and function are
often delayed on that side relative to the better function-
ing kidney. This produces a “flip-flop” pattern; the poorly
functioning side initially has lower uptake but the corti-
cal activity on later images is higher than the better func-
tioning side, which has already excreted the radiotracer.
Clearance Phase
The calyces and pelvis usually begin filling by 3 minutes.
Over the next 10–15 minutes, activity in the kidney and
collecting system decreases. With good function, most of
the radiotracer clears into the bladder by the end of the
study. In some healthy subjects, pooling of activity in the
dependent calyces can result in focal hot spots. Lack of
clearance or overlap of pelvocalyceal structures on the
cortex suggests hydronephrosis. Because areas with
increased activity appear larger, caution must be taken in
diagnosing hydronephrosis on scintigraphic studies.
The normal ureter may or may not be seen. Prolonged,
unchanging, or increasing activity suggests dilatation.
Because peristalsis and urine flow rates cause such vari-
able visualization, care must be taken in diagnosing reflux
into the ureters when activity remains in the kidney. On
these studies, indirect determination of reflux can be
done when ureteral activity persists after the kidneys
have cleared. However, reflux is best detected on a direct
vesicoureterogram ( VCUG) with activity introduced
directly into the bladder via a catheter.
The bladder is normally well seen. Prevoid and postvoid
bladder images evaluate emptying and postvoid residual.
A distended bladder can cause an obstructed pattern. In
a patient with neurogenic bladder or outlet obstruction,
the renal scan is best performed with a urinary catheter in
place. In infants and small children,the bladder may appear
quite large and higher in position than might be expected
when looking at the outline of the child’s body.
Clinical Applications of Renal Scintigraphy
The clinical uses of renal scintigraphy are numerous.
Renal imaging can assess blood flow, relative size, and

A
Activity

0 20 40 60
Time (sec)
B
Figure 8-11 Renal blood flow analysis. A, Sequential 2-second frames show moderately delayed
and decreased blood flow to the right kidney (arrowhead). B, Sixty-second time activity curves
confirm the imaging findings. Initial upslope of the right kidney (R) is delayed compared with the
aorta (A) and left kidney (L).
230 NUCLEAR MEDICINE: THE REQUISITES

functional parameters. Although functional abnormali-
ties can be seen in acute and chronic renal failure, the
pattern is not specific for the etiology ( Fig. 8-12).
Abnormal blood flow can be seen with renal artery
stenosis, thrombosis, avulsion, venous thrombosis, and
infarction. Static cortical imaging is most often done to
evaluate pyelonephritis and some possible masses.
Renovascular Hypertension
More than 90% of patients with hypertension have no
identifiable cause (“essential hypertension”). Renovascular
hypertension ( RVH) accounts for a significant proportion
of those patients with trea causes for their elevated blood
pressure. Although the prevalence of renovascular
hypertension is less than 1% in nonselected patients,
15–45% of patients referred to specialty centers for
refractory hypertension have RVH.
The two main causes of RVH are atherosclerosis and
fibromuscular dysplasia. Not all cases of renal artery
stenosis cause functional renovascular hypertension.
Almost half of normotensive patients over the age of 60
years have atherosclerotic lesions in their renal vessels.

Cts Cts

Left kidney Left kidney

Right kidney Right kidney

Time (min) Time (min)

B
Figure 8-12 Chronic renal failure. A, The Tc-9m MAG3 studies in a patient with elevated
creatinine initially show slow clearance with bilateral cortical retention (top row).The findings are
not specific for the cause of dysfunction. A follow-up study done 6 months later shows no change
and cardiovascular disease was the assumed etiology of the chronic failure (bottom row). B, The
time–activity curves show slight worsening between the two studies, that is, a slower rising curve.
 Genitourinary System 231

In addition, many hypertensive patients have renal artery
stenosis unrelated to their blood pressure, showing no
response to angioplasty or stenting of the stenosis.
When an arterial lesion causes significant vascular steno-
sis in the renal artery or one of its major branches,glomeru-
lar perfusion pressure drops, causing glomerular filtration
to fall. The kidney responds by releasing the hormone
renin from the juxtaglomerular apparatus. Renin converts
angiotensinogen made in the liver to angiotensin I. In the
lungs,angiotensin I is converted by angiotensin converting
enzyme (ACE) to vasoactive angiotensin II, which acts as
a powerful vasoconstrictor. This constriction acts peri-
pherally to raise blood pressure and acts on the efferent
arterioles of the glomerulus to raise filtration pressure,thus
maintaining G F R (Fig.8-13).
If renal blood flow remains low, the kidney will
become scarred and contracted with time. If RVH is pres-
ent, early intervention decreases arteriolar damage and
glomerulosclerosis. This increases the chance for a cure.
It is therefore critical to identify patients who would ben-
efit from invasive therapy such as angioplasty, arterial
stenting,or surgery. However,physical exam and lab tests
are generally nonspecific. It is not practical or desirable to
subject all suspicious patients to invasive angiography. It
is not practical or desirable to subject all suspicious
patients to invasive angiography.
Noninvasive imaging tests play an important role in
evaluation of the kidneys. However, IVP has false-positive
and false-negative rates of approximately 25% for the
detection of RVH. Conventional radionuclide studies using
Tc-99m DTPA and Tc-99m MAG3 have sensitivity and
specificity similar to IVP. Abnormal function is nonspecific
and can be seen in other disease states. CT angiography
and MR angiography are able to identify arterial stenosis
noninvasively, but do not take into account the functional
effects of any anatomic lesion.
The development of ACE inhibition renography using
captopril (Capoten) led to a sensitive, noninvasive func-
tional method for diagnosing renovascular hyperten-
sion. ACE inhibitors work by blocking the conversion
of angiotensin I to angiotensin II ( Fig. 8-14). This causes
GFR to fall in RVH patients relying on the compensatory
mechanism to maintain perfusion pressure. Functional
changes can be seen on scintigrams and renograms.
Indication
ACE inhibition renography should be considered for
patients at moderate- to high-risk for renovascular hyper-
tension. This includes patients with severe hypertension,
hypertension resistant to medical therapy, abrupt or
recent onset, onset under the age of 30 or over the age of
55 years,abdominal or flank bruits,unexplained azotemia,
worsening renal function during ACE inhibitor therapy,
and occlusive disease in other beds.
Imaging Protocol
Two radionuclide studies are performed: one with and
one without ACE inhibitor. An example of an ACE
inhibitor renography protocol is listed in Box 8-5.
Radiopharmaceuticals that have been used include
I-131 OIH and Tc-99m DTPA, but now Tc-99m MAG3 is
most commonly used.It is possible to perform both studies

Afferent
arteriole

Renal
Proximal tubule
artery
stenosis
Glomerulus GFR GFR
normal reduced

GFR
Angiotensin II
maintained removed
Efferent
arteriole Angiotensin II
vasoconstriction

A B C

Figure 8-13 Pathophysiology of renin-dependent renovascular disease: pharmacological effect of

captopril. A, Normal glomerular filtration rate (GFR). B, Renovascular hypertension. Because of
reduced renal plasma flow, filtration pressure and GFR fall. Increased renin and resulting angiotensin
II produces vasoconstriction of the efferent glomerular arterioles, raising glomerular pressure and
maintaining GFR. C, Captopril blocks the normal compensatory mechanism and GFR falls.
232 NUCLEAR MEDICINE: THE REQUISITES
Kidney
Liver Lung
Renin ACE
Angiotensinogen
Captopril
Angiotensin I
Figure 8-14 Renin-angiotensin-aldosterone physiology and site of angiotensin-converting
enzyme (ACE) inhibitor (captopril). See text for detail.
using a 1-day protocol. The 1-day protocol involves first
doing a baseline exam using a low-dose of 1 mCi (37
MBq) of Tc-99m MAG3 radiopharmaceutical followed by
a post-ACE inhibitor study using 5-10 mCi of Tc-99m
MAG3 (185–370 MBq).
Alternatively, the two studies are performed on separate
days with 3–5 mCi of Tc-99m MAG3 (111–185 MBq).
A baseline study is only done in those patients with an
abnormal ACE inhibitor study. The radiotracer should be
allowed to clear and decay,so at least 24 hours between the
studies is needed.
Patient preparation involves discontinuing all ACE
inhibitors (2–3 days for captopril and 5–7 days for longer-
acting agents such as enalapril and lisinopril) before the
study. If ACE inhibitors are not discontinued the sensitiv-
ity for diagnosis of RVH is reduced approximately 15%.
Stopping angiotensin receptor blockers and halting cal-
cium channel blockers should also be considered.
Diuretics should be stopped to prevent a dehydrated
condition. Otherwise, most antihypertensives have little
or no effect on the results.
Intravenous access should be obtained as patients with
renin-dependent renovascular hypertension usually expe-
rience a drop in systemic blood pressure after receiving
ACE inhibitors. Administration of fluids may be critical,
especially if the patient is at risk for severe hypotension or
has cardiovascular risk factors such as recent myocardial
infarction or carotid/coronary artery disease.
The loop diuretic furosemide (Lasix) is frequently
administered simultaneously with the radiopharmaceuti-
Adrenal Aldosterone
Salt-water
retention
Increased
blood volume
Hypertension
Peripheral
Angiotensin II
vasoconstriction
Renal
efferent
arterioles
cal. This ensures clearance of the collecting system,
which might otherwise affect visual and quantitative
interpretation. Furosemide acts on Henley’s loop distal to
where Tc-99m MAG3 is secreted. Therefore, furosemide
does not affect the cortical retention or clearance of
Tc-99m MAG3.
Captopril (25–50 mg) is crushed and dissolved in
water. A 1-hour delay is then necessary to allow absorp-
tion to occur. Although this delay adds inconvenience
and variable absorption may occur, captopril is still fre-
quently used. Alternatively, the ACE inhibitor enalapril
(Vasotec), given intravenously at 40 ug/kg up to 2.5 mg
over 3–5 minutes, can shorten the time before imaging to
15 minutes and ensures that consistent ACE inhibitor lev-
els are achieved. A baseline blood pressure should be
recorded. Serial blood pressure measurements should be
taken and the patient should be carefully monitored for
hypotension through the end of imaging.
Image Interpretation
In patients with renin-dependent renovascular hyperten-
sion, the decreased blood flow to the affected kidney is
most often not seen on baseline or conventional renogra-
phy. ACE inhibitors cause a drop in GFR, which leads to
decreased urine flow that can be visualized during the
functional portion of the study as a diminished function.
However, because the kidney could have abnormal func-
tion due to numerous etiologies, a baseline study is done.
In patients with renovascular hypertension,function gen-
erally improves in the absence of ACE inhibitors. In nor-
mal patients and in those with renal disease due to other

Box 8-5 Angiotensin-Converting Enzyme
(Captopril) Renography
Protocol Summary
PATIENT PREPARATION
Liquids only 4 hours before the study
Hydration: 7 ml/kg water 30–60 minutes before the
study or intravenous hydration 10 ml/kg (maximum
500 ml) half-normal saline over 1 hour. Keep vein
open during study in case of hypotension.
Discontinue certain medications:ACE inhibitors (3 days
short-acting, 5–7 days longer-acting agent),
angiotensin receptor blockers; consider stopping
calcium channel blockers and diuretics
RADIOPHARMACEUTICAL TC-99M MAG3
2-day protocol: 3–6 mCi (111–222 MBq)
1-day protocol: use 1–2 mCi (37–74 MBq) for baseline
then 5–10 mCi (185–376 MBq) ACE inhibitor study
IMAGING PROCEDURE
1. Check baseline blood pressure
2. Administer ACE inhibitor
Captopril: adults, 25–50 mg (dissolve in water) orally
1 hour before study; children, 0.5 mg/kg
(maximum 25 mg)
Enalapril 40 μg/kg (maximum 2.5 mg) intravenously
infused over 3–5 minutes
3. Monitor and record blood pressure every 15 minutes
for oral medication and every 5 min for IV enalapril.
A large drop in pressure may require fluids
intravenously
4. Administer 20–40 mg furosemide intravenously at
the time of injection
5. Inject radiopharmaceutical 60 minutes after oral
captopril or 15 minutes after enalapril
6. Imaging protocol is similar to dynamic renography
(see Box 8-4)
etiologies (such as chronic scarring from pyelonephri-
tis), function does not improve on a baseline study com-
pared to the ACE inhibitor examination.
In general, the diagnostic pattern seen will depend
on the type of radiopharmaceutical used ( Fig. 8-15). If
the glomerular agent Tc-99m DTPA is used, the ACE-
inhibitor–induced drop in GFR leads to a marked drop in
radiotracer filtration and uptake. The most common pat-
tern is an overall drop in function seen as slower uptake
and a lower peak ( Fig. 8-16). A tubular-secretion agent
such as Tc-99m MAG3 does not usually show the same
decrease in uptake; tubular uptake and secretion are not
affected by GFR changes from the ACE inhibitor. The
decreased urine flow causes delayed Tc-99m MAG3
washout and the primary finding will be cortical reten-
tion ( Fig. 8-17).
Genitourinary System 233
R
R
L L
0 10 min 0 10 min
A Before Captopril After Captopril
0 10 min 0 10 min
B Before Captopril After Captopril
Figure 8-15 Renovascular hypertension. A, A comparison of
Tc-99m DTPA and Tc-99m MAG3 curves. Left, Baseline DTPA study
shows normal function bilaterally. Right, After captopril, peak and
overall renal function falls on one side from the drop in GFR. B,
The baseline MAG3 study is unremarkable (left) but ACE inhibitor
study shows the effect on renal function as cortical retention
(right).
If bilateral cortical retention is seen, the finding is likely
artifact and not bilateral renal artery stenosis. Among
patients with bilateral cortical retention sent to arteriogram,
roughly two-thirds had no significant stenosis. Dehydration
or hypotension are often the cause ( Fig. 8-18).
When looking at the TAC patterns, the degree of
change is significant. As seen in Fig. 8-19, several response
patterns are possible after ACE inhibitor imaging. It is
critical to look at the degree of change from baseline.
The pattern may change from a normal baseline to
mildly-reduced or markedly-reduced function. A kidney
with decreased function may progress to worsening
grades of function. The greater the change, the more
likely significant renal artery stenosis is present causing
renovascular hypertension.
Quantitation can aid in assessing functional change
seen on the images. With Tc-99m DTPA, a 10% decrease
in relative function (differential function or split renal
function) or a decrease in absolute function (calculated
GFR) greater than 10% is considered “high probability”or
positive; a change of 5–9% is intermediate or borderline
in significance. Although a change in relative (differen-
tial) cortical function may not be seen with Tc-99m
MAG3, if a change of 10% or greater occurs, it is consid-
ered “high probability.”Although no change will be seen
234 NUCLEAR MEDICINE: THE REQUISITES

Figure 8-16 Effects of renovascular hypertension on Tc-99m DTPA. Top,The baseline study
shows mildly decreased function on the left. Bottom, Exam after captopril reveals severe
deterioration on the left kidney with diminished peak and overall function.

in RPF or ERPF calculated with Tc-99m MAG3, the corti-
cal retention will be reflected in an increased 20/peak or
20/3 ratios. An increased time to peak (Tmax or Tpeak).
The kidney must function well enough to actually
show a change in function when ACE inhibitors are
given. Therefore, a small shrunken kidney or one with
poor baseline function can be difficult to interpret. In
general,ACE inhibitor renography is accurate when the
serum creatinine is normal or only mildly elevated (crea-
tinine <1.7 mg/dL).
If the protocol has been properly followed, sensitivity
and specificity of ACE inhibitor renography have been
reported to be approximately 90% and 95%, respectively.
However, the sensitivity and specificity are lower in
patients with elevated creatinine. False positives are rare
but have been reported in patients on calcium channel
blockers.
Reporting
If the ACE inhibitor study is normal, the study should be
read as “low probability” of a renin-dependent renovas-
cular hypertension. This means the posttest chance of
renovascular hypertension is less than 10%. In general
no additional workup is needed. If the 2-day protocol
was used, the patient need not return for a baseline
study. An “intermediate probability” should be reported
in patients with an abnormal baseline study who show
no change after ACE inhibitors. Patients in this group
would include those with ischemic nephropathy (who
often have a unilateral small kidney), bilateral cortical
retention and those showing only small changes in func-
tion. If these patients were considered as positives, the
sensitivity of the exam would remain high but the speci-
ficity for RVH would be lowered to 50–75%.
A “high probability”( Box 8-6) result indicates a greater
than 90% chance of RVH, and indicates these patients are
highly likely to improve with angioplasty or surgery.
These high-probability exams occur when function
markedly worsens on the ACE inhibitor challenge when
compared to the baseline study. Again, in a positive
study,Tc-99m MAG3 will show worsening as marked cor-
tical retention, whereas Tc-99m DTPA will show a fall in
peak function and a decrease in relative function or GFR.
Urinary Tract Obstruction
Uncorrected urinary obstruction can lead to recurrent
infection, diminished function, progressive loss of
nephrons, and parenchymal atrophy. When the upper uri-
nary tract is obstructed, there is backpressure from the
 Genitourinary System 235

R
Counts

Counts

0 8 16 24 0 8 16 24
Time (min) Time (min)
C

Figure 8-17 Positive captopril study with Tc-99m MAG3. A, The captopril study was performed
first. Note prompt symmetrical initial uptake. Over time the left kidney washes out normally, but the
right shows almost all activity remaining in the cortex. B, Following captopril, the scan shows
marked cortical retention in the abnormal right kidney. C, Left: captopril, right: no captopril.The
pattern is confirmed on the time–activity curves and is “high probability”for RAS as a cause for
renovascular hypertension.
236 NUCLEAR MEDICINE: THE REQUISITES

R
Counts

Counts

L L

0 8 16 24 0 8 16 24
Time (min) Time (min)
C
Figure 8-18 Bilateral cortical retention from dehydration. A, Captopril-stimulated study with
Tc-99m MAG3.There is bilateral cortical retention and minimal urinary bladder clearance over
30 minutes. B, Baseline study without captopril shows normal cortical function and good clearance
into the bladder. C, Left, captopril; right, baseline time–activity curves confirm the impression that
bilateral renal artery stenosis may be present.The arteriogram on this patient was normal. Further
investigation revealed the patient had abstained from food and drink for nearly 12 hours before
captopril study but was hydrated for the subsequent baseline exam, and the resulting dehydration
was presumed to cause the false-positive imaging pattern.
 Genitourinary System 237

0 1 2 3 expected. Therefore, a screening test that could noninva-

sively diagnose significant obstruction is very important.
Ultrasonography is a sensitive method of identifying
a dilated collecting system but can not reliably determine
if this is due to significant mechanical obstruction or
merely nonobstructive hydronephrosis (such as from
Counts

reflux, primary megaureter, or a previous obstruction

which has been relieved). Contrast studies such as IVP
and conventional radionuclide renography show overlap-
ping findings that overlap between obstructed and
nonobstructed systems: delayed filling, dilatation, and
decreased washout. Retrograde pyelography, endoscopy
and noncontrast CT scans can often identify the etiology
4 of an obstructed system (such as a ureteral calculus) but
do not provide functional information important in
0 5 10 15 20
management.
Figure 8-19 Renogram pattern changes with ACE inhibitors.
The Whitaker test, first described in the 1970s,
The possible patterns after captopril with progressive worsening
of function (0–4). Normal TAC curves (grade 0) have low remains the standard for diagnosing obstruction by
probability of RVH.As curve grade (1–4) worsens from the normal measuring pressure-flow relationships in the renal
baseline, the likelihood of RVH increases. Grade 1: Peak mildly pelvis. This is an invasive technique involving bladder
delayed but greater than 5 min and with delayed excretion. Grade catheterization and inserting a needle into the renal
2:Very delayed uptake but some washout. Grade 3: Extremely
pelvis under fluoroscopy. Pressure in the renal pelvis is
delayed uptake with no washout. Grade 4: Complete renal failure
where blood pool moves through the kidney without an measured under basal conditions and after perfusion of
extraction phase. a dilute contrast solution (Fig. 8-20). In a dilated but
nonobstructed collecting system, the differential pres-
sure between the kidney and bladder measures less
than 10–12 cm of water. In an obstructed system, perfu-
sion pressure exceeds 15 cm of water and is frequently
much higher.
Box 8-6 Positive ACE Inhibitor Diuretic Renography
Renography Results Diuretic renography is an excellent, noninvasive method
for evaluating patients with suspected obstruction. Serial
TC-99M DTPA diuretic renography can be performed to determine the
significance of a partial obstruction. It is also possible to
Primary positive finding:
Overall drop in function assess effectiveness of stenting or surgical correction of
Quantitation changes: an obstruction. A list of conditions that this test is used
>10% decrease in GFR for is included in Box 8-7.
>10% decrease in differential function In a dilated system, prolonged retention of radiophar-
Other findings: maceutical is seen due to a reservoir effect. A diuretic
Delayed time to peak causes increased urinary flow and will cause a prompt
Increased cortical retention washout in a dilated but nonobstructed system. If
mechanical obstruction is present, the narrowed lumen
TC-99M MAG3
prevents augmented washout; prolonged retention of
Primary positive finding:
tracer proximal is seen and can be quantified on the
cortical retention
time–activity curves. The ability to perform quantitation
Other findings:
Delayed time to peak is an important advantage of functional radiotracer stud-
Decreased differential function less common ies over those done with intravenous contrast dye.
Radiopharmaceuticals
The original agent used for diuretic renography was I-131
OIH. When renal function is normal,Tc-99m DTPA can
pelvis onto the tubules and vessels. Within hours of be used successfully but should not be used in patients
onset, renal blood flow, glomerular filtration, and renal with azotemia. Tc-99m MAG3 has become the agent of
output are decreased. If a high-grade obstruction is cor- choice because of its excellent quality images and high
rected promptly, function can recover fully; however, if clearance rate, even in patients with renal insufficiency
left uncorrected for over a week, only partial recovery is or immaturity.
238 NUCLEAR MEDICINE: THE REQUISITES

Methods
Numerous protocols for diuretic renography exist, and
attempts to standardize methodology have been made by
comparing data from several institutions resulting in sev-
eral consensus papers. Although there is still some vari-
ability among institutions, many general areas have been
agreed on. An example protocol is listed in Box 8-8.
Patient Preparation Patients must be well-hydrated
so fluid is available for mobilization in response to
diuresis and to prevent dehydration from the diuresis.
The bladder should be catheterized in children and
infants, in those who can not void voluntarily, and when
neurogenic bladder or outlet obstruction is present. This
will help prevent retrograde pressure from a full bladder
from slowing the washout and mimicking the pattern of
upper urinary track obstruction.
Diuretic Administration Furosemide is a loop
diuretic that inhibits sodium and chloride reabsorption,
markedly increasing urine flow in normal patients. The
injection is given slowly over 1–2 minutes with an onset

Box 8-8 Diuretic Renography Protocol

Summary

Figure 8-20 Whitaker test. Dilute contrast solution is infused
into a dilated renal pelvis at the rate of 10 ml/min. Pressure
measurements are obtained to evaluate for suspected obstruction.
PATIENT PREPARATION
Hydration as described in dynamic renography protocol
Place Foley catheter in children; consider in adults
If not catheterized, bladder emptying must be complete
before diuretic injection and after 20-min study.

FUROSEMIDE DOSE
Children: 1 mg/kg to maximum 40 mg (may require
more in severe azotemia)
Adults: base dose on creatinine level
Serum Creatinine
Creatinine Clearance Furosemide
Box 8-7 Urological Conditions Studied (mg /dl) (ml) Dose (mg)
by Diuresis Renography
1.0 100 20
1.5 75 40
Ureteropelvic junction obstruction 2.0 50 60
Megaureter: obstructive, nonobstructive, refluxing 3.0 30 80
Horseshoe kidney
Polycystic kidney IMAGING PROCEDURE
Prune-belly syndrome Inject Tc-99m MAG 3–6 mG (111–222 MBq)
Ectopic ureterocele Acquire study for 20 minutes
Urethral valves Slowly infuse furosemide intravenously over 60 sec
Postoperative states Obtain pre-void and postvoid image in uncatheterized
Pyeloplasty patients
Ureteral reimplantation
Urinary diversion IMAGE PROCESSING
Renal transplant ureteral obstruction On computer, draw ROI around entire kidney and pelvis
Obstructing pelvic mass Generate time–activity curves
Ileal loop diversion Calculate a half-emptying time or fitted half-time

of action within 30–60 seconds and a maximal effect
seen at 15 minutes. The reported diuretic administration
times have varied with injection times at 20 minutes after
the radiopharmaceutical (F+20), at the same time as the
radiopharmaceutical (F+0), and 15 minutes before the
study (F−15). A commonly used method is the F+20
furosemide protocol.
Interpretation
In a normal, nondilated kidney, the TAC rapidly reaches
a sharp peak and spontaneously clears rapidly. Furosemide
diuresis accelerates the rate of radiotracer washout in
a normal kidney. If an ROI is placed over the ureter, a tran-
sient spike after diuretic injection indicates passage of
a bolus of accumulated activity from the renal pelvis.
A dilated but nonobstructed system may initially look
like a normal kidney with a steep TAC uptake slope.
However,a sharp peak is not seen and,as the dilated system
fills, the TAC may show continued accumulation or
a plateau 20–30 minutes after tracer injection. After
furosemide infusion, a nonobstructed hydronephrotic kid-
ney clears promptly due to increased urine flow
( Fig. 8-21). An obstructed system, on the other hand, will
not respond to the diuretic challenge;activity will continue
to accumulate or sometimes stays at a plateau ( Fig.8-22).
The distinction between obstruction and nonob-
structed hydronephrosis decreases as the collecting sys-
tem volume becomes larger. In very distended systems,
delayed washout may be seen whether obstruction is
present or not. An “indeterminate”pattern of clearance is
seen with little change on the images and the TAC
(Fig. 8-23, D). This pattern can be explained by the for-
mula: Tt = V/F, where V is the volume of the system,
F is the flow rate, and Tt is the transit time through the
system. According to this formula, as the volume
increases, the transit time lengthens.
If renal function is very impaired, diuretic response to
lasix may be markedly diminished, leading to prolonged
washout even if no obstruction is present. In patients
with azotemia, an increased Lasix dose or the F−15
diuretic infusion may be used. However, even with addi-
tional modifications, it may not be possible to induce suf-
ficient diuresis to exclude obstruction in a poorly
functioning kidney ( Fig. 8-24).
At times, it is useful to quantitate collecting system
clearance half-time or washout half-time (T ⁄ ). Numerous
1
2
methods for calculating the T ⁄ have been described. The
1
2 dure ( Fig. 8-25). Scintigraphy can assess many complica-
activity can be measured when the diuretic is given; then
the length of time it takes to reach half that level can be
used. As this does not account for the delay in furosemide
effect,a more precise method might be to fit a curve to the
steepest portion of the washout time–activity curve. In gen-
eral,a T ⁄ less than 10–15 minutes indicates no obstruction is
1
2
present, whereas values greater than 20 minutes are con-
sidered obstructed. Values between 15–20 minutes fall in
a gray zone or indeterminate range.
Genitourinary System 239
Serial studies may be used to monitor patients with par-
tial obstruction, with previously treated obstruction, or
who are at risk for worsening obstruction. Some situations
where this might be needed include cervical carcinoma
and known partial ureteropelvic junction obstruction.
Also, patients at high risk for functional deterioration from
reflux, such as ileal loop diversions of the ureters, may be
followed with interval studies. Periodic diuresis renogra-
phy can help determine at what point aggressive interven-
tion is needed.
Pitfalls and Limitations
An indeterminate diuretic renogram is neither positive
nor negative and does not rule out obstruction. The
Whitaker test had a similar category. Some nonob-
structed patients with very hydronephrotic kidney col-
lecting systems will have an indeterminate response
because of the reservoir effect.
A limited response to furosemide may also be seen in
infants. Because neonates have functionally immature
kidneys, the lasix renogram may appear falsely
obstructed. A follow-up study may be helpful in patients
diagnosed with neonatal hydronephrosis. In general, sur-
geons prefer to perform surgery when the infant is larger
and the genitourinary tract more mature.
Renal insufficiency poses another problem. A flat
response may be seen with diuretics because function
is inadequate for effective diuresis. A larger dose of lasix
or administering the Lasix well before the exam begins
( F−15 minutes) can be helpful sometimes. However,
serum creatinine may remain normal if impairment is
unilateral. Although serum creatinine is not a reliable
indicator of the level of functional impairment, radionu-
clide calculation of GFR and ERPF are useful. Generally,
function must not be less than 15% of normal. If the
GFR on the affected side is less than 15 ml/min, diuretic
renography is unreliable.
Other problems may affect interpretation. The effect
of a full or neurogenic bladder must always be consid-
ered. A repeat exam after bladder catheterization may be
helpful. Certain conditions, such as a pelvic kidney or
low-lying renal transplant that may remain overlapped
with the bladder, limit interpretation.
Renal Transplant Evaluation
Renal transplantation is a well-established surgical proce-
tions including acute rejection, acute tubular necrosis (or
more properly termed, vasomotor nephropathy), vascular
problems, and obstruction. Many radionuclide methods
have been used to evaluate the renal allograft function
and identify acute rejection (including gallium-67, labeled
leukocytes, and platelets), but these have lacked speci-
ficity. Conventional renal scintigraphy has been widely
used to evaluate function. Although the number of these
examinations has decreased because many kidneys are
240 NUCLEAR MEDICINE: THE REQUISITES

Furosemide (Lasix)

R
Counts

0 8 16 24 0 10 20
Time (min) Time (min)
B
Figure 8-21 Nonobstructed hydronephrosis. Diuretic renogram of a patient with surgically
treated vesicoureteral obstruction of the right kidney. A, Postfurosemide images show that the
retained activity responds promptly to the diuretic with almost complete clearance. B, Time-activity
curve promptly declines with Lasix, signifying no obstruction.

now evaluated by ultrasound and biopsy, radionuclide
evaluation remains an important screening tool.
Kidneys for transplantation come from three sources:
a deceased donor (cadaveric kidney), a living related
donor, or a living unrelated donor. All potential donors
are carefully screened with immunological matching and
undergo functional and anatomic evaluation. Although
cadaveric kidneys are carefully screened and trans-
ported, allografts from living donors generally have the
best prognosis. One-year allograft survival rates are
90–94% for living related donor kidneys and 88–90% for
cadaveric transplants.
 Genitourinary System 241

Cts
Left kidney
Right kidney

Time (min)
B
Cts

Left kidney
Right kidney

Time (min)
D
Figure 8-22 Obstructed hydronephrosis. A, Progressive filling of an enlarged collecting system is
seen on the left, whereas the right kidney clears normally following Lasix administration at 10
minutes. B, Time–activity curves show washout on the right but no overall clearance on the left
consistent with obstruction. C, Diuretic renography images following surgical correction of the left
ureteropelvic junction obstruction show in the same patient hydronephrosis but improved
clearance. D, Postoperative TACs confirm that no significant obstruction remains on the left.

The development of improved immunosuppressive
drug therapy has been critical in the marked increase in
allograft survival in recent years. These drugs act largely
by suppressing the CD4 T-cell activity. Glucocorticoid
steroids remain essential in the treatment and prevention
of rejection. Now most regimens use triple-therapy,
which include not only glucocorticoids and calcineurin
inhibitors (tacrolimus or less often cyclosporin) but also
an antiproliferative agent (such as azathioprine) that pre-
vents mitosis and nonspecifically suppresses lymphocyte
proliferation. Due to the increased risk of acute rejection
in the immediate posttransplant period, anti-interleukin
(IL)-2 monoclonal antibodies may also be given. T-cell
specific monoclonal antibody muromonab-CD3 (OKT3), as
well as numerous other new and investigational agents,may
be used to prevent and treat acute rejection. However, it is
critical to determine precisely when these agents are
needed and how to monitor patients due to potential seri-
ous acute and long-term side effects of these medications.
Multiple problems may be present in a patient who has
undergone renal transplantation. A list of commonly
encountered complications is provided in Table 8-4. These
include vascular problems, rejection, leaks, and ureteral
obstruction from extrinsic impression or effects of implan-
tation. Rejection is commonly described in terms of the
time frame in which it occurs: hyperacute, accelerated
acute, acute, and chronic. Interpretation depends a great
deal on the length of time from transplantation as well as
242 NUCLEAR MEDICINE: THE REQUISITES

Relative activity

Activity
0 5 10 15 20 25 30 0 5 10 15 20 25 30
A Time (min) B Time (min)
Activity

0 5 10 15 20 25 30
C Time (min)

Figure 8-23 Diuresis renography time–activity curves (TACs). In these examples, Lasix is given at
peak collecting system filling (arrows). A, Normal kidney response to diuretic.The short plateau
before further emptying represents diuretic-induced flow just before rapid clearance. B, Dilated
nonobstructed kidney.The slowly rising curve represents progressive pelvocaliceal filling.With
diuretic (arrow), rapid clearance occurs. C, Obstructed kidney. The diuretic has no effect on the
abnormal TAC. D, Indeterminate response. Following diuretic, very slow partial clearance is seen.
This can be the result of an extremely distended system but obstruction is not excluded.

the type of transplant. A baseline study is very useful in
predicting long-term allograft function and assessing acute
complications.
Medical Complications
Vasomotor Nephropathy or Ischemic Nephro-
pathy A common early complication of the trans-
planted kidney is vasomotor nephropathy (also properly
called ischemic nephropathy or delayed graft function).
Although this is still sometimes commonly referred to as
acute tubular necrosis (ATN), this terminology is not an
entirely correct description. ATN is just one possible
component of the process and there may be little if any
destruction of the tubular elements. Vasomotor nephro-
pathy is actually an ischemic response damaging the
kidney at some point prior to completing transplantation.
The functional impact ranges from a mild, rapidly
 Genitourinary System 243

Figure 8-24 High-grade vesicoureteral junction obstruction secondary to tumor. A, Flow study
shows very decreased perfusion to the left kidney (arrowhead). B, Dynamic sequential images at
5-minute intervals show only a thin rim of cortex with poor uptake (open arrowhead) and a large
photopenic collecting system consistent with hydronephrosis. Diuresis renography would not be
useful as no tracer enters the collecting system.

resolving disorder to a slower recovery or total anuria
(nonoliguric ATN). The severity of damage increases if the
time between the donor’s death and transplantation is
prolonged. Vasomotor nephropathy is most common in
cadaveric kidneys, occurring up to 50% of the time.
However, it can be seen in grafts from living donors (5%),
especially if the surgery was complicated. Therefore, the
renal function will be impaired at the time of transplant
and should recover spontaneously beginning after
transplantation and continuing over 1–2 weeks (Fig.8-26).
Hyperacute Rejection Careful HLA immunologic
matching has eliminated hyperacute rejection. This is
244 NUCLEAR MEDICINE: THE REQUISITES
Figure 8-25 Renal transplant surgery. For technical reasons, the
initial allograft is usually placed in the right iliac fossa.The vessels
are attached end to end for the artery and side to side for the vein.
The ureter is usually directly implanted into the recipient’s
bladder. After an initial failure, a second graft is usually placed in
the left. If a pancreas is transplanted simultaneously, the pancreas is
usually placed on the right.
the result of preformed antibodies in the recipient’s
circulation from a major histocompatibility or blood
group mismatch causing renal vasculature thrombosis.
The classic story is of the surgeon witnessing the kidney
turning blue in the operating room after performing the
vascular anastamosis. When this occurs, the allograft
cannot be salvaged.
Accelerated Acute Rejection Accelerated acute
rejection is also uncommon but occurs in patients with
antibodies already in their system,as in sensitization from
pregnancy or from multiple blood transfusions. The
patient presents with clinical signs of acute rejection at
an earlier time than would be expected. When rejection
develops in the first 3–5 days following transplantation,
accelerated acute rejection should be suspected. Unlike
hyperacute rejection,accelerated rejection often responds
to therapy.
Acute Rejection Acute rejection (AR) is a relatively
frequent transplant complication. Although it can occur
at any time, AR does not typically occur until at least 5–7
days after transplantation and is most common in the first
3 months. Typically, the patient becomes desensitized
to the allograft over time, and AR is rarely seen after
1 year in a patient taking appropriate immunosuppres-
sive therapy.
AR may be caused by two immunologic pathways:
T cells or humoral antibodies. With either of these
immunological mechanisms, the scintigraphic pattern
is similar. Allograft vasculature is affected unlike with
vasomotor nephropathy (Figs. 8-27 and 8-28). Arteritis,
microinfarcts, and hemorrhage occur in the parenchyma
along with lymphocytic infiltration. Clinically, it is com-
mon for the patient to be febrile and the allograft to be
swollen and painful. Laboratory values include rising cre-
atinine and an elevated sedimentation rate.
Chronic Allograft Nephropathy Previously
referred to as chronic rejection, chronic renal allo-
graft nephropathy (CR) is a cumulative, delayed, and
irreversible process. Vascular constriction, chronic
fibrosis, tubular atrophy, and glomerulosclerosis from
immunologic and nonhumoral etiologies occur. Over
months to years, this fibrosis causes cortical loss and
decreased function. Dilatation of the collecting system
may occur as the cortex thins. Risk factors for early
development (less than 1 year) include damage from
early ischemic injury (severe ATN), prior severe rejection,
subclinical rejection, and long-term calcineurin-inhibitor
therapy. There is no effective therapy available for CR and
it is now the leading cause of graft failure given improved
therapy and early transplant survival.
Immunosuppressive Drug Toxicity Another
important cause of allograft dysfunction is drug
nephrotoxicity of therapeutic drugs. Renal allograft drug
toxicity was classically described in relation to
cyclosporin and was often related to high levels of the
drug and early administration. Currently, cyclosporine
toxicity is seldom seen. It has been largely replaced with
other agents and, when prescribed, the lower levels of
the drug are safer. It is important to remember that
similar changes can be seen with other antirejection
agents.
Surgical Complications
Vascular Occlusion Arterial thrombosis is a rare
postoperative complication of transplantation, occurring
in less than 1% of patients. It is usually related to kinking
or difficulty performing the anastamosis with a short
donor artery. No flow or function is seen in this situation.
Renal vein thrombosis may be seen in native kidneys as
a complication of septicemia, dehydration, or nephrotic
syndrome. These patients can be managed most often
medically and function may recover. In the transplanted
kidney, renal vein thrombosis can occur as a postopera-
tive complication or related to autoimmune problems.
The thrombus leads to infarct and hemorrhage. However,
unlike the native kidney,there are no venous collaterals to
drain the kidney and the effect is often severe with loss of
the allograft.
Arterial Stenosis Postoperative renal artery stenosis
should be suspected in patients who develop hyper-
tension following transplantation surgery. Stenosis may
 Genitourinary System 245

Table 8-4 Complications after Renal Transplantation

Usual time of occurrence

Complication postoperatively Comments

PRESURGICAL INSULT
Vasomotor nephropathy (ATN) Minutes to hours Cadaveric transplants; semicaler spontaneous
resolution

AUTOIMMUNE: REJECTION
Hyperacute rejection Minutes to hours Preformed antibodies irreversible
Accelerated rejection 1–5 days History of prior transplant or transfusion
Acute rejection After 5 days, most common during first Cell mediated, responsive to treatment
3 months
Chronic allograft nephropathy Months to years Humoral, irreversible, inevitable
(chronic rejection)
Cyclosporin toxicity Months Reversible with drug withdrawal

SURGICAL
Urine leak/urinoma Days or weeks
Hematoma First few days
Infection First few days
Lymphocele 2–4 months

VASCULAR
Renal artery stenosis After first month
Vascular occlusion
Infarcts
Renal
Obstruction (extrinsic mass, stricture Days, months, years
or calculi)
Vesicoureteroreflux

occur in up to 10% of patients. Although the anastomotic
site is the most common site, stenosis in other locations
(including the iliac artery) must be considered. The
imaging protocol is similar to the renovascular hyper-
tension work up for native kidneys described in Box 8-6.
The only difference is the anterior positioning of the
camera over the allograft. Interpretation criteria are
the same as for native kidneys.
Urinary Leak Necrosis of the ureteral anastomosis in
the immediate postoperative period can result in urinary
leakage. Although other imaging modalities, such as
ultrasound, can identify fluid in the pelvis, nuclear
medicine techniques are better able to specifically
identify the source.
Ureteral Obstruction Although uncommon,ureteral
obstruction may be caused by kinking of the ureter,
extrinsic mass compression (e.g.,hematoma,lymphocele),
intraluminal obstruction (from blood clot or calculus), or
periureteral fibrosis. Some degree of collecting system
dilatation without significant mechanical obstruction is
often seen from postoperative hematomas and seromas.
Hydronephrosis from obstruction must be differentiated
from dilatation caused by reflux. This can be done with
diuretic renography. Ureteral obstruction often resolves
spontaneously.
Lymphoceles Because the transplanted kidney has
no lymphatic connections, lymphoceles can form in the
transplant bed. This may occur in up to 10% of transplants
and is seen typically 2–3 months posttransplant. These
are only clinically important if they impinge on ureter or
vasculature.
Methods
Renal allograft evaluation is performed using the
dynamic scintigraphy protocol with Tc-99m MAG3 listed
in Box 8-4, with the exception of positioning. The patient
is imaged anteriorly with the camera centered over the
allograft in the lower pelvis. It is useful to include at least
some of the native kidneys in the field of view as they
may contribute to overall function. Some portion of the
bladder should be seen, and the entire bladder is
included on prevoid and postvoid images.
Numerous semiquantitative and quantitative methods
have been described. ERPF, GFR and Tc-99m MAG3 clear-
ance are all good indices of renal function. More com-
monly, semiquantitative techniques such as the 20/3 ratio
described in the dynamic renography section are used as
246 NUCLEAR MEDICINE: THE REQUISITES

Cts

Time (min)
B
Cts

Time (min)
D
Figure 8-26 Vasomotor nephropathy (ATN). A, Baseline images obtained 24 hours after
transplantation reveal prompt uptake but significant cortical retention. B, Baseline TAC shows a
steep upslope which goes against vascular involvement and poor overall clearance confirming the
diagnosis of vasomotor nephropathy. C, A follow-up exam shows resolution of the abnormal
findings and relatively good function, which is confirmed on the TAC curves in D.

a measure of radiopharmaceutical transit and cortical
retention.
In general, a study should be done within a day or two
of transplantation to obtain a picture of the baseline level
of function. This baseline is critical in accurately assess-
ing any problems that occur later and improves detection
of subtle abnormalities. The protocol can be modified to
answer any question that arises. If concern for renal
artery stenosis exists, the ACE inhibitor protocol is used.
The diuretic renography protocol can be followed when
hydronephrosis develops or obstruction is suspected.
Delayed images over the course of 1–2 hours can help
clarify the etiology of fluid collections and assess possi-
ble urine leaks.
Interpretation
Because the medical and surgical complications
described previously occur at certain times, renal trans-
plant scintigrams must be interpreted with the age of the
transplant as well as the clinical context (including physi-
cal symptoms, laboratory values, and current medica-
tions) in mind. The type of allograft (cadaveric or living
related donor transplant) is especially important and
needs to be considered when evaluating vasomotor
nephropathy or the expected level of renal function.
 L L
Counts

Counts

0 8 16 24 0 8 16 24
F Time (min) Time (min)

Figure 8-27 Acute allograft rejection.Tc-99m MAG3 images of a right iliac fossa cadaveric
transplant. Good baseline blood flow (A) and good function (B) are seen at the end of first week.
Two days later, the patient developed fever, allograft tenderness and elevated serum creatinine.
Repeat blood flow is diminished (C) and functional images show cortical retention (D).These
findings are consistent with acute rejection. E, Perfusion time activity curves show initial good
blood flow (left) which falls on the follow up study (right) due to acute rejection. F, The TACs over
the 25-minute study show adequate baseline function (left) and that deteriorates at the time of
rejection (right).
248 NUCLEAR MEDICINE: THE REQUISITES

Renogram
59669

44752
Transplant

29835

14917

0
B 0 7 14 20 27

32849 Renogram

Transplant

24637

16425

8212

0
D 0 7 14 20 27

46506 Renogram

34880
Transplant

23253

11627

F 0 7 14 20 27

Figure 8-28 Time course of acute transplant rejection. Baseline functional images (A) and TACs
(B) are unremarkable. Two months later, images show delayed uptake and cortical retention (C).
This is confirmed on the TAC (D). Function improves with immunosuppressive therapy on the
follow-up scan (E) and time–activity curve (F) 1 week later.

A baseline study is essential in the immediate postopera-
tive period to be able to interpret any abnormality seen.
During the perfusion phase, the transplanted kidney
normally becomes the “hottest” structure within 2–4
seconds of appearance of activity in the adjacent iliac
artery following a good bolus of radiopharmaceutical.
The images and functional time–activity curves normally
show a prompt peak and rapid clearance. The expected
level of perfusion and function will typically diminish
over the years as discussed later.
A living related donor allograft will usually function
better and clear radiotracer more rapidly than a cadaveric
allograft. Therefore,“normal” function is somewhat rela-
tive. Because the level of function will vary from kidney
to kidney as well as over time, a baseline study done in
the first 24–48 hours is extremely helpful. If rejection is
suspected later, it may not be diagnosed properly with-
out the baseline comparison.
In vasomotor nephropathy, scintigraphic images show
normal perfusion but poor function with delayed corti-
cal clearance and decreased urine excretion (see Fig.
8-26). Vasomotor nephropathy is seen on the baseline
study done in the first 24–48 hours following surgery,
most often in a cadaveric allograft, and resolves over 1–3
weeks. Although the degree of impairment varies, no
urine output is seen in severe cases. Severe cortical
retention or function that does not rapidly improve on
serial studies has strong negative prognostic implications
with few such kidneys surviving over 6 months. Because
it usually resolves spontaneously, it is suspicious if func-
tion does not improve or if it worsens. There may be
a new ischemic or nephrotoxic insult or, most likely,
acute rejection may be developing.
Nephrotoxic effects of immunosuppressive therapy
may demonstrate a pattern of prompt uptake and
delayed clearance similar to vasomotor nephropathy.
This used to be frequently seen due to cyclosporine toxi-
city and can still occur, although it may be due to other
agents given the currently used low cyclosporin doses.
The time frame of the exam allows these two processes
to be differentiated. Vasomotor nephropathy is seen
immediately whereas time is required for nephrotoxicity
or true ATN to occur.
If hyperacute rejection was ever to be imaged, there
would be essentially no perfusion and a photopenic
area would be seen where the kidney should be on sub-
sequent functional images. This pattern of absent flow
and function can be seen occasionally, but it is due to
severe vascular compromise such as renal artery throm-
bosis or renal vein thrombosis ( Fig. 8-29, A). It should be
noted that renal vein thrombosis in a transplanted kidney
appears differently from the usual appearance in native
kidneys ( Fig. 8-29, B). Native kidneys can rely on venous
collaterals until function often recovers, but these collat-
erals are absent in a transplanted kidney.
Genitourinary System 249
The classic dynamic imaging pattern of acute rejec-
tion is decreased perfusion with delayed uptake. Tc-99m
MAG3 will also show cortical retention. Because AR
affects small renal parenchymal vessels, the allograft
may show diminished blood flow on the perfusion por-
tion of the exam. Because dynamic perfusion imaging is
difficult to do, it is sometimes necessary to assess perfu-
sion based on the early portion of the TAC for the entire
functional part of the study. The delayed radiopharma-
ceutical uptake is seen with a shallow upward slope of
the TAC and an early patchy image. The kidney shows
a prolonged parenchymal transit with cortical retention
( see Figs. 8-27 and 8-28) and may appear enlarged.
Table 8-5 compares acute rejection and vasomotor
nephropathy.
In the immediate postoperative period, AR may be
superimposed on a patient with previously identified
vasomotor nephropathy. In these cases, the cortical
retention of vasomotor nephropathy on the baseline
exam does not resolve as expected or shows worsening
function on serial exams. Obviously, this means a base-
line study is extremely helpful. Although the uptake
slope of the functional curve is usually normal with
vasomotor nephropathy and more shallow with AR, it
may be technically difficult to differentiate the two com-
plications on the basis of differences in perfusion. In
addition, both conditions show cortical retention on the
renogram. However, the later time course of AR com-
pared with vasomotor nephropathy will allow them to
be differentiated if a baseline study has been done. Of
course, other problems such as infection, renal artery or
vein thrombosis, and leak must be excluded.
In CR, the blood flow and function images may ini-
tially appear normal. Often, the changes of CR may only
be revealed through quantitative means such as ERPF and
GFR measurement. Quantitation will show decreased
function with a downward trend over time on serial
examinations (Fig.8-29). As the CR worsens,parenchymal
retention is seen, although this is in a poorly functioning
kidney with all functional parameters appearing abnor-
mal. The cortex appears patchy and less intense than
normal and clearance is delayed. Although mild to mod-
erate cortical retention is often seen in CR, more marked
retention is likely due to another process. When a scan is
ordered in a kidney older than 1–2 years, it is important
to remember that AR seldom occurs if the patient is ade-
quately immunosuppressed. Usually, significant cortical
retention is a sign of true ATN.
The nephrotoxic effect of immunosuppressive therapy
such as cyclosporine A has a similar appearance to AR.
The functional portion of the study shows slow uptake,
cortical retention, and decreased function. A biopsy may
be needed to differentiate changes from drug toxicity and
other chronic allograft diseases from AR or ATN,especially
when blood levels of the drug are not elevated.
250 NUCLEAR MEDICINE: THE REQUISITES

Cts

ERPF: 576 mL/min

B Time (min)

Cts

ERPF: 207 mL/min

D Time (min)

ERPF: 124 mL/min

F Time (min)

Figure 8-29 Chronic renal allograft nephropathy (formerly called chronic rejection). Baseline
scan (A) and baseline time–activity curve (B) 24 hours after transplantation reveal good uptake and
clearance. Note the presence of a Foley catheter draining the bladder and some function in native
kidneys at the top edge of the image. Images done over 1 year later are unremarkable (C), although
some decline is suggested on the time–activity curve (D).A steady decline in ERPF is characteristic
and often the only way to diagnose the early stages of disease. Four months later as creatinine
continued to rise, the scan shows a visible decline in function (E), also seen on the curve (F).
 Genitourinary System 251

Although vascular complications are rare, they must

Table 8-5 Comparison of Acute Rejection and be suspected in the anuric patient. Renal artery occlu-
Vasomotor Nephropathy sion leads to absent perfusion and a photopenic defect
on the functional portion of scintigraphy. Although
Renal renal vein thrombosis in native kidneys has a variable
Baseline Follow-up transit
scan scan Perfusion time
appearance depending on severity and stage of resolu-
tion, the classic appearance is of an enlarged kidney
Acute rejection Normal Worsens ↓ ↑ with intense cortical retention initially; deteriorating
Vasomotor Abnormal Improves Normal ↑ function and a patchy nephrogram are possible later. In
nephropathy a transplanted kidney, there are no venous collaterals to
(ATN)
drain the kidney so the impact is more severe, causing

Figure 8-30 Renal vein thrombosis. A, Renal vein thrombosis in a native kidney. The I-131 01H
scan reveals poor uptake and delayed clearance in the left native kidney which improves on a follow-
up scan 4 months later. B, Images from a renal allograft with renal vein thrombosis.The radionuclide
angiogram demonstrates no perfusion to the transplanted kidney. C, Dynamic images acquired
immediately after the flow study and sequentially every 5 minutes show a photopenic defect
(arrowheads) resulting from nonviable allograft causing attenuation but having no uptake (D).
252 NUCLEAR MEDICINE: THE REQUISITES
the kidney to have a similar appearance to renal artery
occlusion.
A postoperative leak may occur at the anastomotic site
or due to rupture of an obstructed allograft. A progressive
accumulation of activity outside of the urinary tract may
be seen on dynamic scintigraphic images. A slow leak
may be seen as a photopenic defect from a nonradiola-
beled urinoma (Fig. 8-31). If Tc-99m DTPA is being used,
delayed imaging at 2 hours or later may be needed to
detect increasing activity in this fluid collection.
Delayed images with Tc-99m MAG3 may be misinter-
preted as showing a leak due to the normal bowel activ-
ity appearance over time. Other fluid collections such
as lymphoceles and hematomas may be noted as fixed
pararenal, photon-deficient areas.
An obstructed allograft may present with hydronephro-
sis or diminished urine output. Diuretic renography can
be useful in evaluating suspected obstruction in a similar
manner as in native kidneys. It is important that no AR is
present and that function is adequate to respond to the
diuretic (ERPF >75 ml/min).
MEASURING RENAL FUNCTION: ERPF
AND GFR
Clinical assessment of renal function using blood urea
nitrogen (BUN) and serum creatinine is a relatively crude
process, as a significant decrease in function must occur
before changes in BUN and creatinine are seen. Also,
changes in metabolism and diet, differences in muscle
mass, and certain medications can affect creatinine lev-
els. Occasionally, creatinine clearance is measured and is
more accurate. However, the 24-hour urine collection
makes it difficult to use.
In the laboratory, accurate quantification of GFR and
ERPF is possible with substances that are not radioac-
tive. Because it is freely filtered at the glomerulus but is
neither secreted at the tubules nor reabsorbed, the mole-
cule inulin is the model for GFR measurement. If a sub-
stance has complete first-pass extraction through the
kidney, it can measure RPF.Approximately 90% of PAH is
extracted and has been used to estimate RPF; however,
as it is incompletely cleared, the term ERPF is used.
Inulin and PAH are technically demanding to use, requir-
ing continuous infusion to achieve a steady state; its
clearance is then measured through multiple blood and
urine samples.
Nuclear medicine techniques have evolved utilizing
analogs of inulin and PAH. Different combinations of
plasma sampling and imaging techniques are available.
These nuclear medicine techniques provide simple and
accurate methods to quantify function.
Effective Renal Plasma Flow
The radiolabeled forms of hippuric acid, I-131 OIH and
I-123 OIH, have been used to quantify ERPF.The urinary
clearance of I-131 OIH is approximately 85% of PAH.
The difference is attributed to higher plasma protein
binding and differences in tubular transport. Currently,
Tc-99m MAG3 has replaced the hippuran agents in clin-
ical use. Because Tc-99m MAG3 does not undergo
glomerular filtration and has a slightly lower level of
tubular secretion, it has a reported roughly 60% that of
hippuran. However, extensive studies have been done
that prove ERPF values obtained with Tc-99m MAG3
are accurate once corrected for the different extraction
fraction.
Although the most precise ERPF techniques involve
multiple blood samples, researchers have studied clear-
ance patterns to determine the best times for sampling
so that single blood sample protocols can be done. By
waiting for a time point when the radiopharmaceutical
has fully distributed in the body and blood pool, any
decrease in activity should reflect renal clearance.
A blood sample at 44 minutes has shown to be repro-
ducible within 19 ml/min.
Glomerular Filtration Rate
Many radiopharmaceutical analogs of inulin have been
used to evaluate glomerular filtration.Tc-99m DTPA and
I-125 iothalamate (Glofil) are available in the United
States for calculating GFR. In the past, variable protein
binding levels due to impurities in Tc-99m DTPA led to
inaccuracies in GFR calculations.With careful monitoring
to ensure a low level of protein binding near 1%,Tc-99m
DTPA is an excellent radiopharmaceutical for GFR
measurement.
Cr-51 EDTA is widely utilized for GFR evaluation in
many countries but is not available in the United States.
The long half-life (27.7 days) and high-energy gamma
emissions (320 keV) necessitate a low dose not suitable
for gamma-camera imaging. However, delayed scintilla-
tion well counter measurements can be done accurately,
adding to convenience. It closely approximates inulin
clearance without the protein binding problem that can
occur when using DTPA.
Like ERPF determinations, multiple blood sample
techniques were first used to calculate GFR. Single- and
two-blood sample techniques were then developed.The
single-blood sample technique is reliable and most
widely used. However, although the single-sample tech-
nique is the most practical, the blood sampling must be
done much later than with ERPF, most commonly at
3 hours, making it much less convenient.
 Genitourinary System 253

Normal GFR values are approximately 100 ml/min

for adults. The results are accurate unless function is
markedly diminished (GFR <30 ml/min). In infants,
renal function reaches adult levels by 2 years of age.
Pediatric GFR values should be corrected for body sur-
face area.
Camera Technique
Often, an imaging department is not set up to perform
the scintillation counting and wet lab work needed to
calculate GFR and ERPF using blood sampling tech-
niques. Camera based methods require no blood sam-
pling and only a few minutes of imaging time (Fig. 8-32).

Figure 8-31 Postoperative urinary leak. Rapid leakage resulted from disrupted surgical
anastomosis. Note accumulation of radiotracer just inferior to the transplant but superior and lateral
to the bladder. No bladder filling is seen.

30 cm

A
Figure 8-32 Gamma camera technique for quantitation of glomerular filtration (GFR). A, One-
minute image of the Tc-99m DTPA syringe acquired before and after injection, 30-cm distance from
the center of the collimator. B, After injection, 15-sec/frame images are acquired for a total of 6
minutes. C, Kidney and background regions are selected on the images to obtain counts. After
correcting for background and attenuation, the net renal cortical uptake as a percentage of the total
injected dose is determined.
254 NUCLEAR MEDICINE: THE REQUISITES
Precise adherence to protocol is necessary for these
techniques, and they are more prone to error than the
blood sampling methods.
In order to perform a camera-based GFR calculation,
a small known dose of Tc-99m DTPA is counted a set dis-
tance from the camera face to determine the count rate
before injecting it into the patient.The actual adminis-
tered dose is then corrected for the postinjection resid-
ual in the syringe and serves as a standard. If the dose is
too large, it may overwhelm the counting capabilities of
the system and lost counts would affect accuracy,causing
overestimation of GFR.
After injection, images are acquired for 6 minutes.
ROIs are drawn around the kidneys and the counts are
background subtracted. Attenuation of the photons
caused by varying renal depth is corrected using formuli
based on patient weight and height.The fraction of the
standard taken up by the kidneys in the 1–2.5 minute or
2–3 minute frames is correlated with the GFR measured
by one of the accepted standard methods (such as multi-
ple blood sample, single blood sample, or less accurately
by creatinine clearance).A similar camera based approach
can be used for ERPF calculation.
It should be noted that any method for measuring
function may not be “accurate” measurements of func-
tion, but they are proven to be precise. Because these
results are highly reproducible in any one patient,
patients can be followed over time using this method.
RENAL CORTICAL IMAGING
It is often difficult clinically to distinguish upper urinary
tract infections (UTI) from lower tract UTI. However, the
long-term complications and therapeutic implications of
renal parenchymal infection are very different from those
of lower urinary tract disease. Infection and reflux can
lead to cortical scarring, renal failure, and hypertension.
In most instances today, lesions of the renal cortex are
evaluated with the structural imaging modalities ultra-
sound, CT, and sometimes MRI. However, there are spe-
cific instances when nuclear medicine cortical imaging
adds vital information. Most commonly this involves sus-
pected pyelonephritis in the pediatric patient. Occa-
sionally, cortical scintigraphy is used to differentiate
a prominent column of Bertin seen on ultrasound from
a true mass.Although either Tc-99m DMSA or Tc-99m glu-
coheptonate could be used,Tc-99m DMSA is preferred
because of its superior resolution.
Acute Pyelonephritis
Acute pyelonephritis usually results from reflux of
infected urine.The clinical diagnosis of acute pyelonephri-
tis based on fever, flank pain, and positive urine cultures
is unreliable and especially difficult in infants.Therefore,
recurrent infections often occur and lead to significant
damage and scarring.This process is a significant cause of
long-term morbidity, causing hypertension and chronic
renal failure.The need for a noninvasive test to diagnose
acute pyelonephritis is clear.
Originally, contrast IVP was used. However, IVP was
found to be insensitive, in addition to the risk associated
with intravenous contrast. CT can often identify the
inflammatory change in the kidney, as can radiolabeled
white blood cells and gallium-67 citrate. However, these
tests are not suitable for frequent use, especially in
children. Ultrasound is widely used to assess the kidney
and is generally considered an essential part of the
pyelonephritis workup. However, sonography is relatively
insensitive to the inflammatory changes of acute
pyelonephritis,as well as the residual cortical defects and
scars. Reported ultrasound sensitivities range from
24–40% for pyelonephritis and are approximately 65%
for the detection of scars.
Cortical scintigraphy with Tc-99m DMSA is signifi-
cantly more sensitive then sonography. Sensitivities for
acute pyelonephritis are difficult to determine as Tc-99m
DMSA itself is considered the gold standard. Most fre-
quently, cortical scanning is done in children with acute
pyelonephritis. It may also be performed as part of the
workup of patients with vesicoureteral reflux who have
no evidence of active pyelonephritis.
Method
With Tc-99m DMSA, dynamic imaging is not performed
because background clearance is slow and only a small
percentage of the radiotracer is cleared by the kidney.
Only delayed cortical planar or SPECT imaging is acquired.
An example protocol is discussed in Box 8-9.
Planar imaging usually requires at least both posterior
and posterior oblique views. A pinhole collimator or
converging collimator provides magnification and
improved resolution. High-resolution SPECT affords
excellent resolution, although a multiple head camera
with high resolution collimators is preferable. In addi-
tion, children may require sedation for SPECT as they
must be completely still for the entire acquisition.With
Tc-99m GH, similar delayed-imaging protocols are used,
although dynamic imaging can be acquired first.
Cortical Scan Image Interpretation
The normal Tc-99m DMSA scan should show a homoge-
neous distribution throughout the renal cortex. The
shape of the kidney is variable, as is the thickness of the
cortex.The upper poles may often appear less intense due
to splenic impression on the cortex, fetal lobulation, and

Box 8-9 Renal Cortical Imaging Protocol
Summary
RADIOPHARMACEUTICAL
Tc-99m DMSA: adult, 5 mCi (185 MBq); child, 50 mg/kg
(minimum dose of 600 uCi)
Tc-99m GH: adult 15–20 mCi (555–740 MBq); child, 200
mCi/kg
INSTRUMENTATION
Planar camera: large-field-of-view gamma camera,
parallel-hole collimator for differential calculation.
Pinhole collimator for cortical images. Converging
collimator may be used for adults.
SPECT: dual- or triple-head preferred
IMAGING PROCEDURE
Patient should void before starting
Inject radiopharmaceutical intravenously
Image at 2 hours after injection.
Acquire parallel collimator images for 500k on anterior
and posterior views for differential calculation.
Pinhole images acquire for 100k counts per view.
Position patient so each kidney is imaged separately,
but the camera is same distance from patient.
Quantify function as geometric mean (square root of the
product of anterior and posterior counts)
SPECT
Camera: dual-head or triple-headed with low-energy
ultra-high resolution collimator
Matrix: 128 × 128
Zoom: as needed
Orbit: noncircular body contour, rotate 180 degrees, step
and shoot 40 views/head, 3 degrees/stop, 40-sec/stop
Reconstruction: 64 × 64 Hamming filter with high cutoff
Smoothing kernel
Attenuation correction
attenuation from liver and spleen.The central collecting
system and medullary regions are photon deficient as
Tc-99m DMSA tubular binding occurs in the cortex.The
columns of Bertin will show radiopharmaceutical uptake
and may appear quite prominent. In a study performed to
differentiate prominent column of Bertin from true mass,
the Tc-99m DMSA scan will show radiotracer uptake in
a column of Bertin but not in a mass caused by tumor.
Areas of cortical tubular dysfunction from infection or
scar present as cortical defects. This may be caused by
localized mass effect and edema from the inflammatory
process, as well as by actual tubular dysfunction and
ischemia.A tumor will also present as a defect because
cortical scanning is not specific. Therefore, comparison
with ultrasound is advisable.
Genitourinary System 255
Infection may present as a focal ill-defined lesion or as
a multifocal process (Fig. 8.33). A diffuse decrease in
activity may also be seen; if it occurs without volume
loss, it would suggest a diffuse inflammatory process.
Scars would be expected to have localized, sharp mar-
gins and may occur in a small kidney with associated cor-
tical loss. However, it is extremely difficult to tell an area
of acute inflammation from a scar without serial images,
particularly in patients with clinically silent infections.
In the workup of acute abnormalities due to acute
pyelonephritis, serial scans may also be useful to assess
the extent of recovery. In general, it is advisable to wait
6 months from the acute infection to allow recovery. The
acute inflammatory changes in pyelonephritis will usu-
ally resolve over time (nearly 40%) or significantly
improve (44%). Any defect persisting after 6 months
should be considered a chronic scar.
Multiheaded SPECT is more sensitive than planar tech-
niques for detection of small defects because of its better
contrast resolution. However, specificity may be some-
what lower because some apparent cortical defects actu-
ally may be caused by normal variations such as fetal
lobulation and splenic impression. Experience with
either technique can give excellent clinical results.
Clinical Applications
Radionuclide Cystography
Radionuclide cystography was introduced in the late 1950s
to diagnose vesicoureteral reflux (VUR) and is widely
accepted as the technique of choice for the evaluation and
follow-up of children with UTIs and reflux. The radionu-
clide method is more sensitive than contrast-enhanced cys-
tography for detecting reflux and results in considerably
less radiation exposure to the patient. In many centers,
contrast-voiding urethrocystography is reserved for the ini-
tial workup of male patients to exclude an anatomical
cause for reflux,such as posterior urethral valves.
Vesicoureteral Reflux
Untreated reflux and infection are associated with subse-
quent renal damage, scarring, hypertension, and chronic
renal failure. VUR occurs in approximately 1–2% of the
pediatric population. In patients with acute pyelonephri-
tis,VUR is present in approximately 40% of patients. In
untreated patients,VUR is responsible for 5–40% of end-
stage renal disease in patients under 16 years of age and
5–20% of adults less than 50 years old.
VUR is caused by a failure of the ureterovesical valve.
The normal ureter passes obliquely through the bladder
wall and submucosa to its opening at the trigone. As
urine fills the bladder,the valve passively closes,preventing
reflux. If the intramural ureteral length is too short in
relation to its diameter or if the course is too direct, the
valve will not close completely and reflux results. As
256 NUCLEAR MEDICINE: THE REQUISITES
Figure 8-33 Acute pyelonephritis.Tc-99m DMSA study in an 11-year-old child using a pinhole
collimator. A, Multiple cortical defects are seen, particularly in the upper pole. B, Follow-up study
obtained 6 months later, after appropriate antibiotic therapy, shows resolution of most defects.
a child grows,the ureter usually grows in length more than
in diameter, resulting in decreased reflux and eventual
resolution in 80% of patients. Spontaneous resolution by
2–3 years of age is seen in 40–60% of VUR diagnosed pre-
natally. Even among patients with severe reflux, resolu-
tion occurs spontaneously in 20% within 5 years.
Renal damage is more likely in patients with severe
rather than mild or moderate grades of reflux. Reflux by
itself is not pathological; that is, sterile low-pressure
reflux does not cause renal injury. The intrarenal reflux
of infected urine is required for damage to develop. In
patients without reflux, pyelonephritis has been pre-
sumed to be secondary to etiologies such as fimbriated
bacteria, which can climb the ureter.
Antibiotic therapy is the first line of defense. Renal
scaring has been decreased from 35–60% in untreated
patients down to 10% with therapy. The goal of therapy
is to prevent infection of the kidney until reflux resolves
spontaneously. However, antibiotics do not completely
protect the kidney from infection and scar. Therefore,
patients must be carefully monitored, and serial Tc-99m
DMSA scans may be helpful.
VCUG screening is recommended for siblings of
patients with reflux. It must be remembered that reflux
and pyelonephritis may be clinically silent. These siblings
are at an increased risk of approximately 40% for VUR.
Methodology
Indirect radionuclide cystography can be performed as
part of routine dynamic renal scintigraphy done with
Tc-99m DTPA or Tc-99m MAG3. The child is asked to not
void until the bladder is maximally distended. When the
bladder is as full as can be tolerated, a pre-voiding image
is obtained. Then dynamic images are acquired during
voiding. A postvoid image is obtained once voiding is
done. Although this test has an advantage because the
bladder is not catheterized, upper-tract stasis often poses
a problem for interpretation. In addition, the indirect
method cannot detect the reflux that occurs during blad-
der filling (20%).
Direct radionuclide cystography is the most commonly
used method for diagnosing reflux. It is done as a three-
phase process with continuous imaging during bladder
filling, micturition, and after voiding. Besides diagnosing
reflux, this procedure can quantitate postvoid bladder
residuals.
The protocol for radionuclide retrograde cystography
and residual bladder volume calculation is listed in
Box 8-10. The high sensitivity depends on rapid and con-
tinuous imaging. Tc-99m sulfur colloid and Tc-99m DTPA
are the radiopharmaceuticals most commonly used. Tc-9
9m pertechnetate may be absorbed through the blad-
der, particularly if the bladder is inflamed. A solution of
1 mCi of radiotracer in 500 ml normal saline provides
sufficient concentration.
Dosimetry
The absorbed radiation dose is quite low. A list is pro-
vided in Table 8-6. From 50–200 times less radiation is
 Genitourinary System 257

Box 8-10 Radionuclide Retrograde Cystography Protocol Summary

RADIOPHARMACEUTICAL
Tc-99m sulfur colloid, 1 mCi (37 MBq)

PATIENT PREPARATION
Insert Foley aseptically, inflate balloon, tape to secure
Use clean weighed diaper for infants

POSITION
Patient supine with bladder and kidneys in field of view. Camera under table.

INSTRUMENTATION
Camera: single-head, large FOV, collimator: converging for newborns <1 yr; all-purpose otherwise.
Computer: 64 × 64 word matrix
Filling phase: 10-sec frame for 60 sec; pre-void 30-sec; voiding 2-sec/frame for 120 sec; postvoid 30-sec

IMAGING PROCEDURE
Hang 500-ml bag of normal saline 25 cm above table.
Inject radiotracer into tubing connected to bladder catheter
Filling phase: fill bladder to maximal capacity (age 2+) × 30 = volume instill (ml)
Fill until drip slows or voids around catheter

VOIDING PHASE
Place camera perpendicular to table
Place patient sitting on bedpan with back against camera (infants remain supine and void into diaper)
Deflate Foley balloon, have patient void
Measure urine volume (or weigh diaper to determine output)

INTERPRETATION
Total bladder volume residual postvoid volume and bladder volume at initiation of reflux can be measured.
Voided volume (ml) × Residual counts/min
Residual bladder volume (ml) =
Maximal counts/min − Residual counts/min

delivered to the gonads from the radionuclide method Radionuclide cystography is more sensitive than the
than with contrast cystography. radiographic contrast technique. The radionuclide tech-
nique permits detection of reflux volumes on the order
Image Interpretation of 1 ml. In one study comparing the two techniques, 17%
In a normal study, no radiotracer is seen in the ureters or of reflux events were seen only on the radionuclide
kidneys. Any reflux is abnormal and readily detected by exam. Although radionuclide voiding cystography can
the presence of activity above the bladder (Fig. 8-34). miss low level I reflux due to the adjacent bladder activ-
Reflux grades have been described for radiographic con- ity, it is generally accepted that level I reflux is of little
trast studies ( Fig. 8-35). In this system, criteria used consequence. If a study is negative but clinical suspicion
include the level reflux reaches, the dilatation of the renal
pelvis, and ureteral dilation and tortuosity. However,
anatomic resolution is much lower with scintigraphic Table 8-6 Radiation Dosimetry for Tc-99m
methods and calyceal morphology is not well defined. Retrograde Cystography
A radionuclide grading system would report activity
confined to the ureter grade I reflux, similar to the radio- Organ mrads/mCi or cGy/37 MBq
graphic grade I. A scintigraphic grade II would include
Bladder 18–27
reflux to the renal pelvis and corresponds to the x-ray
Ovaries 1–2
cystography grades II and III. If a diffusely dilated system Testes <1–2
is seen on the scintigrams, it corresponds to grades IV-V Kidneys 0.02–0.4
seen with contrast cystography.
258 NUCLEAR MEDICINE: THE REQUISITES

is high, refilling the bladder will improve sensitivity. This gical emergency. Scintigraphy can confirm the clinically
is not routinely done, however. suspected diagnosis of torsion and direct the patient to
surgery while preventing unnecessary exploration of
patients with epididymitis as a cause for the pain.
SCROTAL SCINTIGRAPHY

It has been possible to scintigraphically evaluate blood Radiopharmaceutical

flow to the acutely painful scrotum since the early 1970s.
Tc-99m pertechnetate is the radiopharmaceutical used
Although scrotal scintigraphy is a very sensitive and spe-
for testicular scanning. It serves as a blood flow and blood
cific tool for the differentiation of acute testicular torsion
pool (soft tissue extracellular space) marker. Tc-99m
from acute inflammatory causes, it is now of historical
pertechnetate is rapidly available from a molybdenum
interest. Ultrasound is clearly the method of choice for
generator and requires no time to perform labeling.
evaluation of the painless scrotal mass.
Testicular viability after torsion of the spermatic cord
depends on the length of time between the onset of
pain and surgical correction. Atrophy may occur after as Methodology
little as 4–6 hours after onset and is inevitable by 10 Patient preparation involves giving an oral thyroid-block-
hours after torsion. Therefore, testicular torsion is a sur- ing medication such as potassium perchlorate. This will

Figure 8-34 Vesicoureteral reflux. A, During the filling phase, reflux is seen first on the right,
then bilaterally in B. On voiding, the left side clears better than the right. Reflux is seen in the renal
pelvic region bilaterally from grade II-III reflux.
Continued

Figure 8-34, cont’d.
I II III IV V
Figure 8-35 Vesicoureteral grading system ( International
Reflux Study Committee). I, Ureteral reflux only. II, Reflux into
ureter, pelvis, and calyces without dilatation. III, Mild to moderate
dilatation/tortuosity of ureter and calyceal dilatation. IV, Moderate
dilatation and tortuosity of ureter and moderate dilation of the
renal pelvis. The angles of the fornices obliterated but the
papillary impressions maintained. V, Gross dilation and tortuosity
of the ureter and gross dilation of the renal pelvis and calyces.
Papillary impressions no longer visible in most calyces.
Genitourinary System 259
reduce radiotracer thyroid uptake normally seen with
this agent. However, scanning should not be delayed for
this medication. Rapid imaging is essential,and the entire
exam should only take 15–20 minutes.
Correct positioning is extremely important. A marker
should be placed on the thigh of the symptomatic side to
ensure the correct side is identified. The testicles may be
supported in a scrotal sling with the penis taped out of the
field of view. A rubberized lead marker can be placed
behind the testes to decrease background activity from the
thighs. A lead strip marker may be placed on the median
raphe on later delayed images to better separate right and
left testes,as they are often asymmetric in size and rotated.
Image Interpretation
Normal Pattern
On flow images, the iliac arteries should be seen simulta-
neously and should appear symmetric. Because flow to
the testicles through the spermatic cord and testicular
artery is relatively low, only a low-grade diffuse activity
may be seen without clear-cut visualization of the sper-
matic cord vessels. The static soft tissue phase images
260 NUCLEAR MEDICINE: THE REQUISITES
show low-grade symmetric activity, usually somewhat
less than the thigh.
Bladder activity accumulating over time and penile
blood flow can lead to difficulties with interpretation if
care is not taken. The later images are better saved for
marker placement rather than for interpretation.
Acute Testicular Torsion
The scintigraphic findings in acute testicular torsion
depend on the time that has elapsed since the acute
event. In early torsion, within a few hours of onset, flow
images may show no significant asymmetry, although
occasionally a “nubbin sign” may be seen from activity in
a small portion of the obstructed proximal spermatic ves-
sels. The soft tissue phase images decreased activity may
be seen in the region of the involved testicle (Fig. 8-36).
Delayed Torsion
Later in the course of torsion, beyond the point when tes-
ticular salvage might occur, the image findings change dra-
matically. This stage has been described as “missed
torsion,” although the term delayed torsion is preferred.
A halo of increased activity develops from hyperemia to
the dartos through the pudendal arterial supply to the scro-
tum. The testicle itself appears as a central area of photon
deficiency (Fig. 8-37). Over time the halo effect becomes
more prominent,resulting in a “donut sign.” The donut sign
pattern is not specific and is seen in testicular abscess.
Figure 8-36 Acute testicular torsion. A, Minimal asymmetry (arrowhead) is seen on flow images
(often no flow changes are seen). B, Blood pool images show decreased activity on the right (open
arrowhead) from acute testicular torsion.
Acute Epididymitis
Bacterial epididymitis and epididymoorchitis usually
occur in late adolescence and early adulthood coincident
with the onset of sexual activity. The scintigraphic find-
ings are dramatically different from acute or delayed tor-
sion. The early dynamic phase shows markedly increased
activity in the spermatic cord vessels,and increased activ-
ity is also seen on the delayed soft tissue images (Fig.8-38).
Therefore, it must be remembered that inflammation and
infection produce hyperemia seen as asymmetric blood
flow and soft tissue uptake.
Torsion of the Testicular Appendage
Torsion of one of the testicular appendages is common
from 7–14 years of age. Although painful, it is not serious
and resolves spontaneously. Management is conservative
and nonsurgical. On physical exam, a “blue-dot” sign may
be seen on transillumination and a nodule may be palpable.
Radionuclide studies may be normal or may show focal
low-grade increased activity at the site of inflammation.
Other Disorders
Scrotal scintigraphy is not recommended for the assess-
ment of abscess, varicoceles, hydroceles, or testicular
masses. These are better assessed by ultrasound.
Reactive hydroceles are often present on scrotal scinti-
grams and present as crescent shaped areas of decreased
activity. Varicoceles may show dramatically increased
Figure 8-37 Delayed testicular torsion in a patient with pain for over 24 hours. A, Two-second
frames show increased flow to the painful left scrotum. B, Blood pool images show a halo pattern
on the left consistent with delayed torsion.The testicle was not viable at surgery.

Figure 8-38 Acute epididymitis in a patient with acute pain in the left testicle. A, Increased flow
is seen to the lateral left scrotum. B, Blood-pool images have a similar pattern consistent with
epididymitis.The symptoms resolved over the course of a week on antibiotics.
262 NUCLEAR MEDICINE: THE REQUISITES
activity late in the venous structures. An abscess will
show intensely increased flow and a “donut sign” not
unlike a missed torsion.
SUGGESTED READING
Blaufox MD,Aurell M, Bubeck B, et al: Report of the radionu-
clides in nephrology committee on renal clearance. J Nucl Med
37: 1883-1890, 1996.
Dubovsky EV, Russell CD, Bischof-Delaloye A, et al: Report of
the radionuclides in nephrology committee for evaluation of
transplanted kidney (review of techniques). Semin Nucl Med,
175-188, 1999.
Gates GF: Glomerular filtration rate: estimation from frac-
tional renal accumulation of Tc-99m DTPA (stannous). AJR 138:
565-570, 1982.
O’Reilly P, Aurell M, Britton KE, et al: Consensus on diuretic
renography for investigating the dilated upper urinary tract.
J Nucl Med 37: 1872-1876, 1996.
Pieppsz A, Blaufox MD, Gordon I, et al: Consensus on renal cor-
tical scintigraphy in children with urinary tract infection.
Semin Nucl Med, 160-174, 1999.
Prigent A, Cosgriff P, Gates GF, et al: Consensus report on quality
control of quantitative measurements of renal function
obtained from the renogram: International Committee from the
scientific committee of radionuclides in nephrology. Semin
Nucl Med, 146-159, 1999.
Russell CD, Dubovsky EV: Reproducibility of single-sample
clearance of Tc-99m-mercaptoacetyltriglycine and I-131-
orthoiodohippurate. J Nucl Med 40: 1122-1124, 1999.
Taylor A, Nally J, Aurell M, et al: Consensus report on ACE
inhibitor renography for detecting renovascular hypertension.
J Nucl Med 37: 1876-1882, 1996.
 9
CHAPTER Oncology

Gallium-67 Tumor Imaging Method

Chemistry and Physics Dosimetry
Pharmacokinetics Tumor Therapy
Dosimetry Monoclonal Antibody Imaging and Therapy
Methodology Background
Image Interpretation Radiolabeling
Clinical Applications Clinical Imaging Applications
Hodgkin’s Disease and Non-Hodgkin’s Lymphoma Colorectal Cancer
Malignant Melanoma Ovarian Cancer
Hepatocellular Carcinoma Prostate Cancer
Lung Cancer Lung Carcinoma
Head and Neck Cancer Monoclonal Antibody Therapy of Lymphoma
Abdominal and Pelvic Tumors Y-90 Ibritumomab Tiuxetan (Y-90 Zevalin)
Soft Tissue Sarcomas I-131 Tositumomab (I-131 Bexxar)
Thallium-201, Technetium-99m Sestamibi, Palliation of Bone Pain
and Technetium-99m Tetrofosmin Tumor Imaging Phosphorus-32 (P-32)
Radiopharmaceuticals Strontium-89 (Sr-89)
Thallium-201 Chloride Toxicity
Technetium-99m Sestamibi Samarium-153 (Sm-153)
Technetium-99m Tetrofosmin Re-186 HEDP
Dosimetry Lymphoscintigraphy
Methodology Melanoma
Clinical Applications Breast Cancer
Breast Cancer Radiopharmaceuticals
Bone and Soft Tissue Tumors Methodology
Thyroid Cancer Melanoma
Kaposi’s Sarcoma Breast Cancer
Other Tumors
Peptide Receptor Imaging There are numerous radiopharmaceuticals available for
Neuroendocrine Tumors tumor imaging. Many of these are not specific for any
Indium-111 Pentreotide (In-111 OctreoScan) particular tumor. Rather, they involve organ-specific
Chemistry and Pharmacokinetics radiotracer binding. For example, a bone scan detects
Accuracy the secondary bone remodeling caused by metastasis.
Methodology Other radiopharmaceuticals target a specific aspect of
Dosimetry a tumor such as a receptor or an antibody directed
Image Interpretation against a tumor antigen. The type of localization for sev-
Tc-99m NeoTect eral radiopharmaceuticals is listed in Box 9-1.

263
264 NUCLEAR MEDICINE: THE REQUISITES
Box 9-1 Overview of Tumor-Imaging
Radiopharmaceuticals
ORGAN-SPECIFIC
Cold Spot Imaging
Thyroid imaging: Iodine-123,Technetium-99m
pertechnetate
Liver imaging:Technetium-99m sulfur colloid
Hot Spot Imaging
Brain scans:Technetium-99m DTPA,Technetium-99m
glucoheptonate
Bone imaging:Technetium-99m MDP, Technetium-99m
HDP
NONSPECIFIC
Gallium-67 citrate
Thallium-201 chloride
Technetium-99m sestamibi
Technetium-99m tetrofosmin
Fluorine-18 fluorodeoxyglucose (FDG)
TUMOR-TYPE SPECIFIC
Iodine-131: papillary-follicular thyroid cancer
Iodine-131 MIBG: neural crest tumors (adrenal medulla
tumor imaging)
Iodine-131 NP-59: adrenal cortical tumor imaging
Technetium-99m HIDA: hepatocyte origin tumors
Radiolabeled monoclonal antibodies against tumor
surface antigens
Indium-111 OncoScint:colorectal and ovarian cancer
Technetium-99m CEA-SCAN: colorectal cancer
Indium-111 ProstaScint: prostate cancer
Technetium-99m Verluma: small cell carcinoma of
the lung
Radiolabeled peptides
Indium-111 OctreoScan: somatostatin receptor
imaging of neuroendocrine tumors
Tc-99m NeoTect: somatostatin receptor imaging of
lung carcinoma
This chapter discusses tumor-imaging radiopharma-
ceuticals not covered in other chapters (Table 9-1).
Some of these are nonspecific tumor-imaging agents
such as gallium-67 (Ga-67), thallium-201 (Tl-201),Tc-99m
sestamibi (MIBI), and Tc-99m tetrofosmin. Another non-
specific imaging agent, F-18 fluorodeoxyglucose (F-18
FDG), is covered separately in Chapter 10.
Among the agents with specific tumor uptake, the
radiotracer may bind to a cell membrane receptor or to
a surface antigen. The receptor binding of indium-111
(In-111) OctreoScan is based on its similarity to the pep-
tide hormone somatostatin. Monoclonal antibody bind-
ing to specific tumor surface antigens is found in agents
such as In-111 OncoScint (colorectal and ovarian can-
cer), Tc-99m CEA-Scan (colorectal cancer), In-111
ProstaScint (prostate cancer),Tc-99m Verluma (small cell
lung carcinoma), and In-111 Zevalin or I-131 Bexxar
(lymphoma).
Two topics related to tumors will be discussed: pal-
liation of metastatic bone pain with bone-seeking radio-
pharmaceuticals and lymphoscintigraphy. Sentinel lymph
node detection has become an important tool in the man-
agement of melanoma and breast cancer.
GALLIUM-67 TUMOR IMAGING
Gallium-67 (Ga-67) was initially evaluated as a bone-
imaging agent. However, it gained widespread use as
a tumor imaging agent (Fig. 9-1). In 1969, Ga-67 was
first used clinically for the detection of Hodgkin’s dis-
ease. Subsequently, Ga-67 showed a high degree of sensi-
tivity for non-Hodgkin’s lymphoma,metastatic melanoma,
and hepatocellular carcinoma. Although Ga-67 is taken
up by numerous other tumors (such as those of lung,
head and neck, and soft tissue sarcomas), its clinical role
in these diseases has been limited.
The evaluation of Hodgkin’s and non-Hodgkin’s lym-
phoma has been the most common clinical use for Ga-67
use. For years, Ga-67 and computed tomography (CT)
were used together in the staging, restaging, and post-
therapy monitoring of lymphoma. Over time, CT
became the primary staging modality, although Ga-67
provided superior results for restaging and therapy
response questions. Currently, the role of Ga-67 is
increasingly limited in the evaluation of lymphoma as the
use of F-18 FDG positron emission tomography (PET) in
lymphoma expands.
Chemistry and Physics
Ga-67 does not have optimal physical characteristics for
scintigraphic imaging. It decays by electron capture and
emits several gamma rays: 93, 185, 288, and 394 keV
(approximately 100, 200, 300, and 400 keV). Usually, the
lower three peaks are used for imaging because of their
higher abundance. Neither the high- nor low-energy
photons are well-suited for gamma camera imaging. The
low-level photons are subject to a higher scatter fraction
and the high-energy photons penetrate the collimator
septa, which degrades image quality.
Pharmacokinetics
Gallium is a group III element in the periodic table (Fig.
1-1) with a biological behavior similar to iron. The biodis-
tribution in the body is complex. As an iron analog, it is
 Oncology 265

Table 9-1 Physical Characteristics of Tumor-Imaging Radionuclides

Photopeaks

Chemical or Physical Principal Percent Dose in

Radionuclide pharmaceutical half-life (hr) mode of decay keV abundance mCi (MBq)

Gallium-67 Citrate 78 Electron capture 93 41 10 (370)

185 23
300 18
394 4
Thallium-201 Chloride 73 Electron capture 69–83 94 3 (111)
Tc-99m Sestamibi 6 Isomeric transition 140 88 25 (925)
Tetrofosmin 25 (925)
CEA-SCAN 30 (1110)
In-111 Pentetreotide (OctreoScan) 67 Electron capture 173 90 6 (222)
OncoScint 247 94 5 (185)
ProstaScint 5 (185)

transported in the blood by iron transport proteins, espe-

Figure 9-1 Normal gallium-67 distribution.Whole body scan of
a 50-year-old woman obtained 72 hours after injection. Highest
uptake is normally seen in the liver, followed by bone and marrow.
Uptake is seen in the colon, lacrimal glands, nasopharyngeal
region, and breast.
cially transferrin. The kidney excretes 15–25% of the
administrated dose within the first 24 hours. From that
point on, however, the bowel is the major route of excre-
tion. Total body clearance is slow, with a biological half-
life of 25 days. Two days after injection, about 75% of the
administered dose remains in the body.
The pharmacokinetics and distribution of Ga-67 cit-
rate are not optimal from an imaging standpoint. It has
slow background clearance, so imaging is usually delayed
until 48 or 72 hours after injection. The considerable
uptake normally seen in liver, bone, and bone marrow
makes tumor detection difficult in or near these organs.
The excretion into bowel and the slow large-bowel clear-
ance create problems detecting abdominal tumor
uptake. The use of laxatives and further delayed imaging
up to 7 days may help overcome this problem.
Ga-67 biodistribution and uptake can be altered by
a number of substances (Box 9-2). The effects are
largely seen in the bone marrow, liver, and kidneys.
Elevated serum iron from multiple transfusions or iron
dextran administration competes with Ga-67 for trans-
ferrin binding sites. Chemotherapy and radiation ther-
apy also commonly change the normal pattern of
activity. For example, vincristine administered within
24 hours of Ga-67 injection can depress liver uptake,
whereas interstitial nephritis from chemotherapy
increases renal uptake. The MRI contrast agent
gadolinium may cause decreased tumor uptake. These
changes are reversible. The effects of chemotherapy,
for example, usually resolve 4–6 weeks after the end of
treatment (Fig. 9-2).
266 NUCLEAR MEDICINE: THE REQUISITES

Box 9-2 Factors Altering Ga-67
Biodistribution
INCREASED RENAL UPTAKE
Chemotherapy
Iron overload: multiple transfusions, iron dextran
Renal disease:ATN, acute renal failure
Infiltrative processes: leukemia, lymphoma, amyloidosis
CHF
DECREASED HEPATIC UPTAKE
Chemotherapy
Iron overload
Liver failure
DIFFUSELY INCREASED OSSEOUS UPTAKE
Chemotherapy
Iron overload
Chronic anemia
Gadolinium contrast
Leukemia
Elevated serum aluminum
Tumor localization of Ga-67 relates to increased iron
uptake and binding in cancer cells. An adequate blood sup-
ply is necessary for tracer delivery and increased vascular
permeability plays a role in accumulation. Uptake appears
to predominantly rely on Ga-67 binding to transferrin recep-
tors on tumor cell membranes. Once inside the cell, the
radiopharmaceutical binds to intracellular proteins and
localizes in the lysosomes. Transferrin receptor levels drop
after therapy, leading to decreased Ga-67 accumulation.
Other methods of localization may be important. For exam-
ple, binding to lactoferrin may also be important in certain
tumors such as Burkitt’s lymphoma and Hodgkin’s disease.
Dosimetry
With a typical adult dose of 10 mCi, the large intestine
receives the highest radiation, about 9 rads (9 cGy). The
spleen and bone marrow receive 5–6 rads (5–6 cGy), and
the liver receives 4.6 rads (4.6 cGy). Dosimetry informa-
tion is listed in Table 9-2. It should be noted that Ga-67 is
excreted in the breast milk.
Methodology
Advancements in instrumentation and technique
have resulted in marked improvements in image
quality and tumor detectability compared to the early
years of Ga-67 imaging. Some of these advances include
improved gamma camera resolution, multiheaded
detectors, improved computers, multichannel acquisi-
tion, and single photon emission computed tomogra-
phy (SPECT).
Although 3-5 mCi (111–185 MBq) of Ga-67 is used for
imaging inflammatory processes, 8–10 mCi (260–370
MBq) is used for tumor imaging. The higher administered
dose results in a greater count rate. This allows for delayed
imaging with increased target-to-background ratios and
high-quality planar and SPECT images. Delayed imaging is
needed for bowel activity clearance and increased tumor-
to-background ratio. The higher radiation dose is therefore
acceptable because it improves tumor detectability at very
low risk in these cancer patients.
Bowel cleansing before imaging has been advocated
to minimize the impact of slow bowel clearance.
However, this is controversial. Laxatives may lead to irri-
tation and increased bowel activity. Patients should be
well hydrated and do not need to fast.
Although the exact protocol varies between labora-
tories, planar and SPECT images are typically obtained
48–72 hours after injection. SPECT images should be
done routinely for the chest and abdomen. This
improves sensitivity and localization of abnormal
lymph-node activity. Planar images can be obtained as
long as 7–10 days after injection and SPECT imaging
can be done at 5–6 days. Detection of mediastinal
tumors may require oblique views if SPECT is not
employed. Box 9-3 describes a typical Ga-67 tumor
imaging protocol.

Figure 9-2 Gallium-67 altered biodistribution. A, Posterior view of the lumbar spine in a patient
with lymphoma initially shows normal bone marrow activity. B, Following chemotherapy, the
marrow activity increases, becoming isointense to liver. C, One year later, the biodistribution has
normalized with activity much decreased.
 Oncology 267

Because chemotherapy and radiation therapy before

Table 9-2 Radiation Absorbed Dose for Common Ga-67 imaging can result in altered biodistribution, Ga-67
Tumor-Imaging Radiopharmaceuticals in injection should be postponed at least 3 weeks following
rads/mCi (cGy/3.7 MBq)
chemotherapy. If Ga-67 imaging is needed sooner, such
as re-evaluation as required before the next 2-week cycle
Tc-99m Tc-99m of chemotherapy, Ga-67 injection should be injected at
Organ Ga-67 Tl-201 MIBI tetrofosmin
least 1 week after prior treatment and 48 hours before
Kidney 0.41 1.2 0.067 0.47 the next therapy.
Thyroid 0.5 0.02
Heart wall 0.5 0.017 0.017
Liver 0.46 0.6 0.02 0.017 Image Interpretation
Spleen 0.53
Bone marrow 0.58 Soft tissue background activity can be high, but
Bone 0.44 depends a great deal on body habitus. Delayed imag-
Gallbladder 0.067 0.2 ing at 48–72 hours is routine because time is needed
Testes 0.2 0.5 0.01 0.13 for the high background activity to decrease. Ga-67
Ovaries 0.3 0.5 0.05 0.04 uptake is normally seen in the liver, spleen, salivary
Urinary bladder 0.07 0.07
Large intestine 0.9 0.4 0.2 0.1 glands, bone marrow, and lacrimal glands (Fig. 9-1).
Breasts 0.007 0.013 The liver has the highest uptake of Ga-67, followed by
Total body 0.26 0.23 0.2 0.02 bone, marrow, and then spleen. The kidneys have
intense uptake for the first 24–48 hours after injection.
Target organ in boldface type.

Box 9-3 Gallium-67 Citrate Tumor Imaging: Protocol Summary

PATIENT PREPARATION
Optional bowel preparation

RADIOPHARMACEUTICAL DOSE
Adult dose: 10 mCi (370 MBq)
Pediatric dose: 75–100 μCi/kg (minimum 500 μCi)

INSTRUMENTATION
Camera: Large field of view; dual-headed camera preferable
Collimator: Medium-energy parallel hole
Photopeaks: 20% windows around 93, 184, and 296 keV
Computer acquisition matrix: 128 × 128 byte mode

PROCEDURE
Whole body images initially at 48–72 hr and at 5–10 days as needed
SPECT of chest, abdomen, or both at 48–72 hr and delayed SPECT as needed up to 5–6 days
1. Inject Ga-67 intravenously.
2. Planar imaging: For dual-headed camera, simultaneous anterior and posterior whole body scanning mode requires 30-40
min. For single-headed camera, obtain 500k spot images of anterior chest and equal time for posterior chest, anterior and
posterior abdomen, pelvis, and anterior head. Regions of special interest require 1000k. Image axillae with arms elevated.
3. SPECT:
Camera Single-headed Dual-headed

Collimator: Medium energy Two medium energy

Rotation: 360°
Patient: Supine Supine
Computer acquisition parameters: 64 × 64 matrix 128 images/360° 128 × 128 matrix 120 images/360°,
arc 20 sec/image 60 stops/head 40 sec/stop at 46 hr
Processing: Filtered back-projection or iterative
reconstruction
Attenuation correction: Attenuation correction:
Chest: no Chest: no
Abdomen: yes Abdomen: yes
268 NUCLEAR MEDICINE: THE REQUISITES

Renal uptake should decrease over time and appear
only faintly by 48–72 hours.
Large bowel clearance is variable and can pose inter-
pretive problems. It may be difficult to differentiate
tumor from normal bowel. In addition, tumor may be
masked by high levels of bowel uptake. Further delayed
imaging may help clarify problems. Abdominal tumor
activity remains fixed as the bowel activity.
Uptake is variable in the salivary and lacrimal glands as
well as in the nasal mucosa. Lacrimal gland uptake is due
to lactoferrin binding. Uptake in the parotid and
lacrimal glands, the “panda sign,” suggests Sjögren’s syn-
drome or sarcoidosis (see Fig. 12-4). Radiation therapy
for head and neck cancer may cause increased salivary
uptake that can persist for years.
Female breast uptake varies with hormonal status
(Fig. 9-3) and is particularly high with postpartum lacta-
tion. Oblique, lateral, and SPECT images can separate
breast parenchyma from intrathoracic processes.
Axillary lymph node activity may be missed due to over-
lapping tissues if the patient is not imaged with the arms
up (Fig. 9-4).
Although any intense or asymmetrical nodal uptake is
abnormal, faint hilar uptake may be seen normally and is
common after chemotherapy. Hilar and mediastinal
uptake can be caused by inflammatory processes such as
granulomatous disease. If Ga-67 uptake is seen on a pre-
treatment scan but correlative CT reveals no abnormality,
the uptake is not likely to be of any clinical significance.
Ga-67 accumulates at sites of recent trauma, surgery,
or infection. Correlation with physical exam and history
is very important. Postoperative sites may have
increased activity for 2–3 weeks. Infectious etiologies
such as pneumonia must be considered when abnormal
Ga-67 accumulation is seen.
Multiple factors affect the ability of Ga-67 to detect
a tumor. First, tumor histology plays a key role. The sen-
sitivity of Ga-67 in several tumor types is listed in Table
9-3. Within a tumor type, high-grade tumors are more
likely to accumulate Ga-67 than low-grade tumors.
Lesion size is another important factor. Tumors less than
2 cm in diameter are not reliably detected with conven-
tional planar imaging. In addition, very large masses
such as those larger than 5 cm may be poorly visualized
because of tumor necrosis. SPECT imaging improves
detection of smaller lesions (Fig. 9-5). Because of its
improved contrast resolution, lesions on the order of
1–1.5 cm can be visualized. The sensitivity of SPECT for
tumor detection is 85–96% compared with 69% with pla-
nar imaging.

Figure 9-3 Variable Ga-67 uptake in the female breast. A, The female patient with Hodgkin’s
disease shown in Fig. 9-2 had intense breast activity on the initial scan.This was attributed to breast
feeding which had been recently terminated. B, The activity resolved on a scan performed 6 months
later. An underlying nodule was revealed in the right chest. C, The nodule resolved after the
completion of chemotherapy.

Figure 9-4 Axillary node uptake of gallium-67. A–B, Initial study reveals left axillary uptake
(arrowhead ) only when the arms are elevated. C, A follow-up study 3 months later shows resolution
of the nodal involvement.
 Oncology 269

Table 9-3 Sensitivity of Gallium-67 for Tumor Table 9-4 Hodgkin’s Disease Versus Non-Hodgkin’s
Detection Lymphoma

Tumor Sensitivity (%) Clinical utility Hodgkin’s Non-Hodgkin’s

disease lymphoma
Hodgkin’s disease >90 +++
Non-Hodgkin’s lymphoma 85 +++ Cellular derivation Unresolved 90% B-cell
Hepatocellular carcinoma 90 +++ Reed-Sternberg 10% T-cell
Soft tissue sarcomas 93 +++
SITE OF DISEASE
Melanoma 82 ++
Lung cancer 85 ++ Localized Common Uncommon
Head and neck tumors 75 ++ Nodal spread Contiguous Not continuous
Abdominal and pelvic tumors 55 + Extranodal Uncommon Common
Mediastinal Common Uncommon
Abdominal Uncommon Common
Bone marrow Uncommon Common
Systemic symptoms Uncommon Common
Curability >75% <25%

Initially, HD is usually seen as localized nodal disease in

Figure 9-5 Gallium-67 thoracic SPECT. A 35-year-old man with
Hodgkin’s disease. A, Anterior planar chest image shows low-
intensity uptake in the right hilum (arrowheads). B, SPECT shows
definite hilar uptake (arrowhead ) on sequential coronal chest
sections owing to the improved contrast resolution of SPECT.
Clinical Applications
Hodgkin’s Disease and Non-Hodgkin’s Lymphoma
Hodgkin’s disease (HD) and non-Hodgkin’s lymphoma
(NHL) differ clinically and pathologically (Table 9-4).
the neck or supraclavicular area (Fig. 9-6). Tumor
involvement in the chest occurs in almost two-thirds of
patients. Orderly spread to contiguous lymph nodes is
typical. The Rye classification system describes four his-
tological subtypes (Table 9-5). However, the stage of dis-
ease is the main factor in determining prognosis, not the
histology. The Ann Arbor staging system is described in
Box 9-4. Ga-67 has long played an important role in HD
because staging is so critical.
NHL is a typically multicentric disease (Fig. 9-7) and
80% of patients have abdominal presentations involving
mesenteric and retroperitoneal nodes (Fig. 9-8).
Roughly 25% of patients with NHL will have extranodal
sites and 45% will demonstrate intrathoracic tumor. The
histologic cell type of the tumor is highly predictive of
survival. Therefore, unlike HD where tumor stage is
most important, tumor histology largely determines
patient prognosis in NHL (Box 9-5).
Aggressive combination chemotherapy, with or with-
out radiation therapy, can cure or produce long-term
remission in a large percentage of patients with HD as
well as many with high- and intermediate-grade NHL.
Coincidentally, these are the same tumor types which
typically accumulate Ga-67. Sensitivity for low-grade
tumors, on the other hand, is poor. Low-grade tumors
generally have an indolent clinical course. Ultimately,
these patients relapse, transform to a higher-grade tumor,
and cannot be cured.
Accuracy
Much of the data cited in the literature are older and do
not reflect several factors that improved Ga-67 sensitiv-
ity. For example, lower-dose regimens (3–5 mCi)
were used that are now reserved for inflammatory
processes. In addition, camera technology and the
270 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-6 Hodgkin’s disease in a 25-year-old man with a left neck mass. A, Left lateral view
shows a large region of gallium-67 uptake in left side of the neck with a small nodal focus just
inferior to it. B, Anterior view. Small focus can be seen inferior to mass on left side of neck.There is
also focal uptake in the mediastinum (proven with SPECT).

should be noted that Ga-67 is highly sensitive and the

Table 9-5 Rye Classification of Hodgkin’s Disease
information it reveals often compliments the CT. In
the past, Ga-67 scans helped clinicians determine that
Histological subgroup Incidence (%) Prognosis staging laparotomy was not necessary for HD. One
Lymphocyte predominant 2–10 Excellent problem with Ga-67 staging in lymphoma is that
Nodular sclerosis 40–80 Very good it often does not add information to the CT due to
Mixed cellularity 20–40 Good lower sensitivity in low-grade tumors and in extran-
Lymphocyte depleted 2–15 Poor odal disease (skin, gastrointestinal tract, kidneys, and
testis).
Currently, Ga-67 is most often used to assess response
widespread use of SPECT have improved results. HD is to therapy. It is essential that a pretreatment scan is
generally detected with a sensitivity of 90–95% and obtained to determine if the tumor is Ga-67 avid. Until
specificity approaching 100%. Sensitivity for NHL is recently, Ga-67 imaging was done at the end of a therapy
slightly lower at 85–90%, but specificity is similar to cycle (Fig. 9-9). Ga-67 can accurately stratify patients
that with HD. Sensitivity is much lower for low-grade into high-risk and low-risk prognostic groups in this
tumors, which are detected with Ga-67 only about half process. Any patient with persistent disease has a very
of the time. poor prognosis. At times, these restaging scans reveal
Ga-67 Lymphoma Staging, Restaging, and Response new sites of disease as well (Fig. 9-10).
to Therapy Ga-67 scans done early in the course of therapy, some-
CT is the primary modality for initial lymphoma staging times after a single cycle of chemotherapy or in mid-
with Ga-67 playing a much smaller role. However, it cycle, allow an even better assessment of prognosis than

Box 9-4 Ann Arbor Staging System for Hodgkin’s Disease

STAGE I Involvement of single lymph node region or single extralymphatic site

STAGE II Involvement of two or more lymph node regions on the same side of diaphragm; can also include localized
involvement of extralymphatic site
STAGE III Involvement of lymph node regions or extra lymphatic sites on both sides of diaphragm
STAGE IV Disseminated involvement of one or more extralymphatic organs with or without lymph node involvement
 Oncology 271

Figure 9-7 Non-Hodgkin’s lymphoma.A 67-year-old man with multiple sites of gallium-67 uptake
by tumor both above and below the diaphragm. A–C, Anterior spot views of the head, chest, and
abdomen and pelvis, respectively. D–F, Posterior views of the chest and abdomen and right lateral
view of the abdomen and pelvis, respectively.

images from the end of therapy. Patients with little
response early in therapy have a lower survival rate with
more frequent relapse than an early responder. One
hypothesis for this finding is that cells resistant to first-
line therapy are seen on these early scans. If imaging is
delayed until after therapy is completed, the resistant
cells may be present but not abundant enough to visual-
ize. These cells are more likely to cause a recurrence
than cells completely killed early in therapy. Not only
does early imaging appear to be more accurate than scan-
ning at the end of the chemotherapy cycle, but it is also
a better predictor of survival than other factors such as
tumor stage or CT appearance. Early modification of
treatment protocol may benefit the high-risk patient.
Residual Mass
Another common use for Ga-67 is evaluation of a mass
persisting on CT after therapy. Because CT images the
size and shape of a mass rather than viability, CT cannot
differentiate between viable tissue and fibrotic scar.
Residual radiographic abnormalities are seen in 64–83%
of patients with mediastinal disease and 30–50% of those
with abdominal disease. In the mediastinum, it has been
reported that Ga-67 has a sensitivity of 96% and a speci-
ficity of 80% for residual active tumor. CT, on the other
hand, was 68% sensitive and only 60% specific for tumor.
Malignant Melanoma
Most malignant melanomas and metastases are gallium
avid. Ga-67 has been used to detect metastases and
determine response to chemotherapy or immunother-
apy. The overall sensitivity and specificity for detecting
metastasis are reported to be 82% and 99%, respectively.
F-18 FDG PET is currently playing an increasing role in
clinical care for melanoma.
272 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-8 Gallium-67 abdominal SPECT. A 26-year-old woman with non-Hodgkin’s lymphoma.
A, The anterior (right) and posterior (left) planar images suggest uptake in the spine or prevertebral
region (arrowheads). B, High-contrast SPECT sequential coronal views clearly confirm prevertebral
periaortic node involvement.A defect is seen in the right lobe of the liver from a subcapsular
hematoma as a complication of liver biopsy. C, Computed tomography. Left: Superior cut shows the
large hematoma. Right: The tumor mass is anterior to the spine in a lower cut. D, Three-view SPECT
display shows the tumor to be anterior to the spine, perhaps best seen in the sagittal view.

Hepatocellular Carcinoma Lung Cancer

Although hepatocellular carcinoma is most often seen on
CT as a single mass in the liver, it is frequently multifocal
in patients with cirrhosis and hepatitis C. Because most
hepatomas are gallium avid, Ga-67 has been used to dif-
ferentiate hepatoma from a regenerating nodule
(“pseudotumor”) on CT (Fig. 9-11).
Overall, the sensitivity of Ga-67 for lung cancer has been
reported to be 85–90%. However, gallium scanning is
very limited in its ability to stage patients and determine
operability. F-18 FDG PET has taken over this role.
Uncommonly, Ga-67 has been used to determine the
extent of pleural-based mesotheliomas and to help
 Oncology 273

Box 9-5 Revised European-American

Lymphoma (REAL)
Classification

B-CELL NEOPLASMS
Precursor B-cell neoplasm: B-cell lymphoblastic
lymphoma
Peripheral B-cell neoplasms
Chronic lymphoma or leukemia
Mantle cell lymphoma
Follicular lymphoma
Marginal cell lymphoma
Hairy cell leukemia
Plasma cell myeloma
Diffuse large B-cell lymphoma
Burkitt’s lymphoma

T-CELL AND NATURAL KILLER CELL NEOPLASMS

Precursor T-cell neoplasm: T-cell lymphoblastic
leukemia/lymphoma
Peripheral T-cell and natural killer-cell neoplasms
Chronic lymphoma or leukemia
Large lymphocyte leukemia
Mycosis fungoides
Peripheral T-cell lymphomas
Angiocentric lymphoma
Intestinal T-cell lymphoma

differentiate malignant mesothelioma from benign

pleural thickening.

Head and Neck Cancer

Varying results have been reported for Ga-67 in head
and neck tumors. Sensitivity for tumor detection
ranges from 56–86%. MRI and CT are the primary
modalities for diagnosis. The prognosis is poor for
patients with gallium avid residual masses after treat-
ment. However, PET is better able to stage and restage
head and neck cancer.
Abdominal and Pelvic Tumors
Ga-67 has been used successfully for the detection
of draining nodal metastasis in testicular cancer.
Uptake depends somewhat on histological subtype:
74% sensitive for metastatic embryonal cell carcinoma,
57% for metastatic seminoma, and 25% for testicular car-
cinoma.
Sensitivity of Ga-67 is generally poor for other tumors
in the abdomen and pelvis. Reported sensitivities are:
esophageal cancer, 41%; gastric tumors, 47%; colon can-
cer, 25%; pancreatic tumors, 15%. Similar low sensitivi-
ties have been reported for gynecologic tumors.
Figure 9-9 Non-Hodgkin’s lymphoma: response to therapy.
Resolution of gallium-67 uptake after appropriate therapy in a
patient with non-Hodgkin’s lymphoma. A, A large portal hepatic
mass is seen before therapy. B, After therapy a residual mass was
seen on computed tomography. No Ga-67 uptake is detected,
although a photopenic mass effect appears to be present just
below the liver because of residual nonviable tumor seen as a mass
on CT.
Soft Tissue Sarcomas
Most soft tissue sarcomas will accumulate Ga-67. An
overall sensitivity of 93% has been reported for primary
lesions, local recurrence, and metastatic disease.
Liposarcoma, usually a low-grade tumor, has a high false-
negative rate. A Ga-67 positive site that resolves after
therapy is indicative of a favorable response.
274 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-10 Hodgkin’s disease: response to chemotherapy. A 30-year-old woman with nodular
sclerosing Hodgkin’s disease. A, Whole body gallium-67 scan shows multiple sites of tumor in the
right perihilar and peritracheal regions, anterior mediastinum, and right and left lungs. Note uptake
in the left buttock at the site of injection (small arrowhead ). B, Follow-up scan after a course of
chemotherapy shows resolution of Ga-67 uptake in the chest. New uptake in the stomach is
secondary to gastritis, best seen in posterior view (large arrowhead ). Gastric localization was
confirmed by SPECT.

Figure 9-11 Hepatocellular carcinoma: SPECT gallium-67 and Tc-99m sulfur colloid.Transaxial
(A) and coronal (B) SPECT slices. SPECT was performed with an aging single-headed rotating
gamma camera.The Tc-99m sulfur colloid liver spleen slices (top) show a large defect (arrowheads)
in the posterior aspect of the right lobe. In comparable sections, the Ga-67 study (bottom) shows
increased uptake (arrowheads) in the same area, consistent with the suspected tumor.

THALLIUM-201, TECHNETIUM-99M
SESTAMIBI, AND TECHNETIUM-99M
TETROFOSMIN TUMOR IMAGING
Myocardial perfusion agents, thallium-201 chloride (Tl-
201),Tc-99m sestamibi (Tc-99m MIBI), and Tc-99m tetro-
fosmin are taken up in a large number of benign and
malignant tumors. This section will briefly describe
these agents and some possible applications. The use of
Tl-201 for the detection of brain tumors is discussed in
Chapter 13. The role of Tc-99m sestamibi in parathyroid
adenoma diagnosis is covered in Chapter 5.
Radiopharmaceuticals
Thallium-201 Chloride
Chemistry and Physics
Tl-201 is a group IIIA metal in the periodic table (see
Fig. 1-1) It decays by electron capture, emitting character-
istic x-rays ranging from 69–83 keV (94% abundant)
and two gamma rays, 167 keV (10% abundant) and
Figure 9-12 Normal resting thallium-201 distribution. Imaging
started 15 minutes after injection show prominent uptake in the
kidneys, heart, liver, and to a lesser extent, the bowel. Normally, the
thyroid would be prominently seen.This patient has undergone
total thyroidectomy for thyroid cancer.Adherence of Tl-201 to the
arm vein on the side of the intravenous injection is common.
Oncology 275
135 keV (3% abundant) with a physical half-life of
73 hours (Table 9-1).
Pharmacokinetics and Normal Distribution
After intravenous injection,Tl-201 is distributed through-
out the body in proportion to regional blood flow (Fig.
9-12). The heart receives 3–5% of the administered dose,
the liver receives 15%, and the kidneys receive 3.5%.
Cardiac uptake is maximal at 10 minutes and peak
uptake is probably similar for most tumors. Biological
clearance is primarily via the kidneys and, to a much
lesser extent, through the intestines. Total-body clear-
ance is slow, with a 40-hour biological clearance.
Multiple factors play a role in the uptake of Tl-201 by
tumors (Box 9-6). Blood flow is critical for radiotracer
delivery. Tl-201 is then handled by cells as an analog of
potassium. Tumor accumulation largely depends on the
sodium-potassium ATPase pumping system on the cell
membrane. It remains free in the cytosol with minimal
localization in the nucleus or mitochondria.
Technetium-99m Sestamibi
Tc-99m MIBI (Cardiolite, marketed as Miraluma for
breast tumor imaging by Dupont Pharmaceuticals) is
a lipophilic cationic complex (methoxy-isobutyl-isoni-
trile). The technetium radiolabel provides superior
images compared with Tl-201.
Pharmacokinetics and Normal Distribution
Compared with Tl-201, Tc-99m MIBI has less cardiac
uptake (2%) and remains fixed in the heart. MIBI clears
rapidly from the blood and localizes in skeletal muscle,
liver, and kidneys (Fig. 9-13). Hepatic uptake is initially
high with radiotracer then clearing into the biliary sys-
tem and bowel. Subdiaphragmatic tumor detection is
more difficult due to intestinal and urinary clearance.
Mechanism of Tumor Uptake
The cellular uptake of Tc-99m MIBI is related to its
lipophilic properties and appears to lead to passive diffu-
sion into the cell. A strong electrostatic attraction
BOX 9-6 Factors Determining Tumor Cell
Uptake of Thallium-201 and
Technetium-99m Sestamibi
THALLIUM-201 TC-99M SESTAMIBI
Blood flow Blood flow
Tumor viability Tumor viability
Tumor type Tumor type
Sodium-potassium Lipophilic cation
ATPase system
Co-transport system Large negative
transmembrane
potential
Calcium ion channel system
276 NUCLEAR MEDICINE: THE REQUISITES
Figure 9-13 Normal resting technetium-99m sestamibi
distribution. Imaging at 60 minutes after injection reveals
prominent uptake by heart and liver. Hepatobiliary clearance and
gallbladder filling are seen, as is intestinal and urinary clearance.
occurs between the positive charge of the lipophilic
Tc-99m MIBI molecule and the negatively-charged mito-
chondria. Approximately 90% of Tc-99m MIBI is concen-
trated within the mitochondria. Tumors are detected as
abnormal areas of uptake.
A cellular membrane glycoprotein, P-glycoprotein
(Pgp), is responsible for pumping cationic and lipophilic
substances out of the cell, including Tc-99m MIBI and
chemotherapy agents. Malignant cells have increased
expression of the multidrug resistance gene (MDR-1),
which encodes for Pgp. Thus, increased amounts of the
chemotherapeutic drugs are transported out of the
tumor cells and may play an important role in drug resist-
ance. With high levels of Pgp, more MIBI is transported
out of the tumor cells. It has been postulated that
Tc-99m sestamibi clearance might be used as an MDR-1
indicator and thus predictive of chemotherapy efficacy.
Technetium-99m Tetrofosmin
Chemistry and Physics
Tc-99m tetrofosmin (Myoview) is a lipophilic cationic
diphosphine (trans-dioxo-bis) complex. When Tc-99m
pertechnetate is added to tetrofosmin in the presence of
the reducing agent stannous ion, a lipophilic, cationic
Tc-99m tetrofosmin complex is formed.
Pharmacokinetics and Normal Distribution
Myocardial uptake of Tc-99m tetrofosmin is rapid. Like
Tc-99m MIBI, it is not cleared from the myocardium.
Tc-99m tetrofosmin is cleared more rapidly from the
lung, blood, and liver than Tc-99m MIBI, which may be
advantageous for detection of tumors in the inferior
quadrant of the right breast.
Mechanism of Tumor Uptake
The mechanisms of tetrofosmin uptake and sestamibi
uptake are probably similar. Both are lipophilic cationic
complexes and the uptake of both correlates with perfu-
sion, high intracellular levels of mitochondria, and cell
viability. Accumulation and retention in the mitochon-
dria are mediated by the negative potential of the mito-
chondrial membrane. Tetrofosmin is also a substrate for
Pgp. The Na+/K+ ATPase pump is only partially involved
in the cellular uptake of tetrofosmin.
Dosimetry
Tl-201 results in a somewhat higher radiation dose to the
patient than that of the Tc-99m-labeled agents (Table 9-2).
The kidney is the critical organ for Tl-201, receiving
1.2 rads per 3 mCi (0.03 cGy/37 MBq). With a 30-mCi
(1110 MBq) administered dose of the technetium agents
(Tc-99m sestamibi and Tc-99m tetrofosmin), the organs
receiving the largest radiation dose are the large bowel.
The gallbladder receives the next highest dose.
Methodology
The imaging protocol should be modified for the particu-
lar clinical indication. Whole-body planar views,regional
“spot views,” and SPECT imaging are all possible. The
optimal time to begin tumor imaging with these agents is
approximately 5–30 minutes after injection. Specific
protocols are discussed in the next section.
The technetium radiolabeled agents have better imag-
ing characteristics than Tl-201. Thallium is suboptimal
because of its low-energy (69- to 83-keV) mercury x-ray
emission and low allowable administered dose (3 mCi or
111 MBq), which limits photon yield. Because of the bet-
ter dosimetry of the Tc-99m-labeled agents, higher doses
(25–30 mCi) are administered. Thus imaging time can
be shorter and the images better.
Clinical Applications
Breast Cancer
Mammography detects breast cancer with a high degree of
sensitivity (85–90%). However,its positive predictive value
for malignancy is low (20–30%) and thus many women
undergo unnecessary surgical biopsies. Mammography
also has a poor negative predictive value in women who
have dense breasts,implants,or those who have undergone
breast surgery or radiotherapy. The false-negative rate in
this group of patients approaches 30%.
Ultrasonography can differentiate cyst from solid
tumor, but it is often otherwise nonspecific. MRI is very
sensitive for tumor detection and can add diagnostic
information in some cases, but its specificity is not high.

A noninvasive imaging test with high positive and negative
predictive values could obviate the need for surgical
biopsy in many women.
Tl-201 is taken up by adenocarcinomas of the breast.
In a study of 45 patients with breast lesions greater than
1.5 cm, Tl-201 had a sensitivity of 97% for detecting
breast cancer. In that study, fibrocystic disease showed
no Tl-201 uptake. The smallest detectable primary lesion
was approximately 1 cm in diameter. Most of these
patients had palpable lesions.
Because of the better imaging characteristics of
Tc-99m MIBI, studies of its utility for breast imaging were
undertaken. Over 20 studies have been reported. In
1997,Tc-99m MIBI became the first radiopharmaceuti-
cal to be approved by the U.S. Food and Drug Administra-
tion (FDA) for breast imaging.
In a large multicenter trial of 673 patients from 30
institutions, an overall sensitivity of 85% and specificity
of 81% were reported for diagnosis of breast cancer in
patients who had a palpable breast mass or a mammo-
graphically detected lesion. Sensitivity was better for
palpable masses (sensitivity 95%, specificity 74%) than
for nonpalpable lesions (sensitivity 72%, specificity 86%).
Sensitivity was also lower for lesions less than 1 cm in
diameter. Another large, multicenter study of 530
patients with a palpable breast mass reported a sensitiv-
ity of 90% and specificity of 87.5%. In this study, the neg-
ative predictive value was nearly 99%, whereas the
positive predictive value was only 50.8%.
Fibroadenomas are the most common cause for false-
positive studies. The positive and negative predictive
values for axillary node metastatic involvement are
approximately 83% and 82%, respectively.
Methodology
A typical imaging protocol is described in Box 9-7.
Tc-99m MIBI scintimammography is best performed
with the patient lying on a specially designed imaging
table with cutouts that allow one breast to hang depend-
ent with the patient in the prone position. This setup
allows lateral images of each breast without background
activity from the chest wall and heart. Supine images are
obtained for two-dimensional tumor localization. A nar-
row window of 10% rather than the typical 20% window
around the 140-keV photopeak is recommended to mini-
mize table scatter. SPECT has not proven advantageous.
Image Interpretation
Breast tumor scintigraphy should be interpreted in con-
junction with the physical examination, mammography,
and ultrasound if available. An abnormal study consistent
with malignancy will show focal increased uptake in the
region of the palpable or mammographically detected
mass (Fig. 9-14). Diffuse uptake is nonspecific and usu-
ally does not indicate malignancy.
Clinical Role
Although Tc-99m MIBI breast imaging has a limited
clinical role currently, the technique is clearly useful
Oncology 277
BOX 9-7 Scintimammography: Protocol
Summary
PATIENT PREPARATION
None
DOSE
Tc-99m sestamibi 25 mCi (925 MBq)
INSTRUMENTATION
Camera: Large field of view with low-energy all-purpose
collimator; 10% photopeak over 140 keV.
IMAGING PROTOCOL
Position patient prone on table with cutouts so that
breasts hang dependent.
Inject Tc-99m sestamibi intravenously.
Begin imaging 5 min after injection.Ten minutes per
image. Marker images may be shorter.
Prone: Lateral of breast with palpable nodule or
mammographically detected mass. Repeat lateral
image with radioactive marker over palpable nodule.
Lateral of opposite breast.
Supine: Chest, including axilla
for certain subsets of patients. For example, patients
with nondiagnostic mammograms, those with dense
breasts or architectural distortion (e.g., from surgery
and breast implants), and those with fibrocystic
disease who are at increased risk for malignancy may
benefit.
The current false-negative rate of 15% means an unac-
ceptably high number of patients with tumor are not
detected. Continued developments, such as new dedi-
cated breast imaging devices with better camera sensitiv-
ity and image resolution, may increase the role of Tc-99m
MIBI in breast imaging.
Bone and Soft Tissue Tumors
Tl-201 can successfully differentiate malignant from
benign bone lesions. A high degree of correlation has
been found between Tl-201 uptake and response to
chemotherapy (Fig. 9-15). The lack of Tl-201 uptake in
a mass indicates tumor necrosis, and tumor response
results in decreasing Tl-201 uptake. Tl-201 is superior to
both Tc-99m MDP and Ga-67 for imaging of bone and soft
tissue tumors. This is not surprising because uptake of
the latter two radiopharmaceuticals is determined by fac-
tors other than tumor response to therapy, such as bone
repair. Frequently, extensive edema is associated with
the tumor on MRI and may obscure tumor margins.
Better definition of tumor may be possible with Tl-201.
Tc-99m sestamibi has performed similarly to Tl-201 in
evaluating bone primary sarcomas.
278 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-14 Scintimammography. Large palpable breast mass imaged with technetium-99m sestamibi.
A, Laterals of right and left breast show intense focal uptake in right breast near the axilla consistent with
malignancy. B, Anterior view helps localize the uptake to the upper outer quadrant.

Figure 9-15 Thallium-201 uptake in osteosarcoma. A, Tc-99m HDP bone scan shows uptake in
the distal left femur extending into soft tissue medially in a young patient with an osteosarcoma.
B, Tl-201 study shows a pattern of uptake similar to that with Tc-99m HDP, but Tl-201 more clearly
shows soft tissue involvement superomedially.Thallium study demonstrates viable tumor.

Thyroid Cancer
Although iodine-131 (I-131) is the primary agent for
imaging and therapy of differentiated thyroid cancer, it
does not accumulate in poorly differentiated tumors,
anaplastic tumors, and medullary thyroid cancer.
Tl-201 has been used for tumor localization when the
I-131 whole body scan is negative but the patient’s
serum thyroglobulin level is elevated. Some of these
Tl-201 positive patients will still show a response to
I-131 therapy as evidenced by decreased thyroglobulin
after therapy. F-18 FDG is now approved for this same
indication.
Medullary thyroid cancer (MTC) arises from the
parafollicular C-cells and does not accumulate I-131.
MTC is a component of the familial multiple endocrine
neoplasia type 2 (MEN-2) syndrome (Box 9-8). Tl-201
and Tc-99m MIBI accumulate in the primary tumor and
can be used to detect recurrence. However, Tc-99m(V)-
dimercaptosuccinic acid [Tc-99m(V)-DMSA] has been
shown the most sensitive of these agents. The reported
sensitivity of Tc-99m(V)-DMSA has ranged from 50% to
95%. The variability in sensitivity may be the result of
various isomeric forms of the DMSA depending on
the preparation. It is not available in the United States.
F-18 FDG PET is useful in the evaluation of MTC and the
use of In-111 pentreotide (OctreoScan) is discussed later.
Kaposi’s Sarcoma
Tl-201 can be used in the diagnosis of pulmonary dis-
ease in AIDS patients. Kaposi’s sarcoma is Ga-67 nega-
tive but Tl-201 positive. Most other infectious
pulmonary diseases are gallium avid (e.g., Pneumocystis,
atypical and typical Mycobacterium). Tl-201 scinti-
graphy is usually negative in infectious and inflamma-
tory disease.
Other Tumors
A variety of other tumors, such as lung cancer, lym-
phoma, and head and neck tumors, have been imaged
with the thallium and technetium radiopharmaceu-
ticals. However, the clinical role of these agents is
limited.
Box 9-8 Multiple Endocrine Neoplasia
(MEN) Syndromes
LESION MEN-I MEN-IIA MEN-IIB
Pituitary adenoma +
Pancreatic islet cell tumor +
Parathyroid adenoma + +
Pheochromocytoma + +
Medullary thyroid cancer + +
Ganglioneuroma + Figure 9-16
Oncology 279
PEPTIDE RECEPTOR IMAGING
Numerous endogenous peptides that modulate tumor
cell growth and metabolism have been identified. These
peptides include several hormones and growth factors
that interact with receptors on the tumor cell membrane.
Among these are somatostatin, vasoactive intestinal pep-
tide (VIP), tumor necrosis factor, and angiogenesis factor.
This section discusses somatostatin derivatives which are
available for therapy and clinical imaging.
Neuroendocrine Tumors
Neuroendocrine cells derive from embryonic neural
crest cells. They share the ability to synthesize amines
and produce peptide hormones and neurotransmitters.
Neuroendocrine tumors have long been associated
with the inheritable multiple endocrine neoplasia syn-
dromes (MEN syndromes) along with other tumors (see
Box 9-8).
Tumors arising from these cells fall into one of three
categories: (1) neuroendocrine tumors or amine pre-
cursor uptake decarboxylation tumors (APUDomas),
including pituitary adenomas, gastric endocrine tumors
(carcinoid, gastrinoma, insulinoma), pheochromocy-
tomas, medullary thyroid carcinoma, and small cell lung
cancer; (2) central nervous system tumors (astrocy-
tomas, meningiomas, and neuroblastoma); and (3) other
tumors including lymphoma, breast, lung, and renal cell
cancer.
Somatostatin receptors have been identified on many
different cells and tumors of neuroendocrine origin
(Fig.9-16). Somatostatin is a 14–amino-acid long peptide
produced in the hypothalamus, pituitary gland, brain-
stem, gastrointestinal tract, and pancreas. Somatostatin
acts as a neurotransmitter in the central nervous system.
Outside of the brain, it functions as a hormone that
inhibits release of growth hormone, insulin, glucagon,
Neuroendocrine cells and the tumors originating
from each type.
280 NUCLEAR MEDICINE: THE REQUISITES

gastrin, serotonin, and calcitonin. Somatostatin has an
antiproliferative effect on tumors. It appears to play
a role in angiogenesis inhibition and is involved in the
immune function of white blood cells.
Several agents have been developed which readily
bind to somatostatin receptors (Fig. 9-17). Octreotide is
an 8–amino-acid segment of somatostatin that maintains
active binding properties of the native peptide hormone.
Unlike the native molecule, it is resistant to enzymatic
degradation in the body. This is reflected in the half-life
of 2–3 hours rather than 2–3 minutes, as seen with
endogenous somatostatin. Nonradiolabeled Octreotide
(Sandostatin) has been approved by the FDA as a thera-
peutic agent suppressing growth in acromegaly and con-
trolling symptoms in metastatic carcinoid and vasoactive
intestinal peptide.
is rapidly cleared by the kidneys. There is also a low
level (only 2%) of hepatobiliary excretion. At 4 hours
after injection, 10% of the dose is still in circulation; at
24 hours, less than 1% is in circulation. This rapid clear-
ance enhances the target-to-background ratio.
Five different subtypes of human somatostatin recep-
tors (SSTR) have been identified (SSTR1 to SSTR5).
These receptors are expressed to varying degrees on dif-
ferent tumors. This explains the differing sensitivities of
somatostatin receptor imaging radiopharmaceuticals.
The commercially available radiopharmaceutical In-111
pentetreotide binds with high affinity to the SSTR2 and
SSTR5 subtypes, to a lesser extent with SSTR3, and not at
all with SSTR1 or SSTR4. Identifying the specific recep-
tor subtypes on tumors is also important as future thera-
peutic agents are developed to target tumors.

Indium-111 Pentreotide (In-111 OctreoScan)
Octreotide was initially radiolabeled with iodine-123.
However, this is a technically demanding process and
images obtained with this radiotracer are limited by
a significant amount of bowel activity. Improved images
are seen with indium-111 (In-111) labeled octreotide.
The In-111 label allows delayed imaging not possible
with a technetium label. Also, In-111 labeled octreotide
has a low level of bowel activity. This agent, In-111 pente-
treotide or In-111 OctreoScan (OctreoScan, Mallinckrodt),
has been approved by the FDA for imaging of neuroen-
docrine tumors.
Chemistry and Pharmacokinetics
The In-111 pentreotide radiolabeling process involves
complexing octreotide with diethylenetriamine pen-
taacetic acid (DTPA) to bind In-111. In-111 OctreoScan
Accuracy
The accuracy of In-111 pentreotide for diagnosis of vari-
ous neuroendocrine tumors is noted in Table 9-6. Many
of these tumors are small and can easily be missed on
conventional imaging. However, sensitivity for small
lesions (less than 1 cm) is limited. The ability to perform
SPECT in addition to whole-body imaging increases
detectability.
For most neuroendocrine tumors, such as gastrinoma
and carcinoid, the sensitivity is very high. Two excep-
tions are insulinoma and medullary carcinoma of the
thyroid, with only 50% sensitivity. The sensitivity for
pheochromocytoma and neuroblastoma is high (approx-
imately 90%), similar to that obtained with I-131 MIBG
imaging (see Chapter 5). For adults, MIBG scanning is
generally preferred, even though the image quality is
poorer and the radiation dose is higher than with
OctreoScan. The advantage of I-131 MIBG is the higher

Ala-Gly-Cys-Lys-Asn-Phe- Phe D-Phe-Cys- Phe

D-Trp D-Trp
Table 9-6 Accuracy of Indium-111 OctreoScan
s s
s s
in Multicenter Trial
Lys Lys
Cys-Ser-Thr-Phe- Thr Thr-OL-Cys- Thr
Consistent/total
Somatostatin Octreotide Tumor type patients (n)* Percent

Carcinoid 190/237 80
123I Insulinoma 8/11 31
Gastrinoma 40/42 95
D-Phe-Cys- Phe 111In-DTPA-D-Phe-Cys- Phe Glucagonoma 8/11 73
s D-Trp Small cell carcinoma of lung 2/2 100
s D-Trp
s Lys s Pheochromocytoma 9/9 100
Lys
Thr-OL-Cys- Thr Paraganglioma 6/7 86
Thr-OL-Cys- Thr Medullary thyroid carcinoma 12/22 54
Vipoma 6/7 86
I-123 octreotide In-111 pentetreotide Pituitary adenoma 24/30 80
Figure 9-17 Comparison of somatostatin analog octreotide,
iodine-123 octreotide, and indium-111 pentetreotide *Other methods included biopsy, computed tomography, ultrasonography,
(OctreoScan). magnetic resonance imaging, and angiography.

target-to-background ratio and better specificity. An
important disadvantage of In-111 pentreotide is its
persistent high kidney activity, which makes interpreta-
tion of the adjacent adrenal gland more difficult. The
reported sensitivity for other tumors, such as lymphoma
and lung and breast cancer, is about 70% each; however,
clinical utility has not been established.
Methodology
Box 9-9 describes a typical imaging protocol for In-111
OctreoScan. Because Sandostatin competes with In-111
OctreoScan for uptake, the therapeutic agent Sandostatin
is usually discontinued 3–7 days before the study. However,
there are case reports of better visualization of metas-
tases while patients were taking the drug. Imaging is nor-
mally done at 4 and 24 hours. Early imaging at 4 hours is a
dvantageous because bowel activity is absent at this
early time, although the background activity is still high.
The tumor detectability improves on 24-hour images due
to background clearance. Additional images can be done
at 48 hours after bowel cleansing if needed.
Dosimetry
The estimated radiation-absorbed dose to the patient
from OctreoScan is shown in Table 9-7. The spleen
is the target organ, receiving the highest absorbed dose
followed by the kidneys.
BOX 9-9 Indium-111 OctreoScan: Protocol
Summary
RADIOPHARMACEUTICAL
Dose: 6 mCi (222 MBq) In-111 OctreoScan intravenously
PREPARATION
Bowel preparation with laxative and enema; hydration
Discontinue octreotide therapy 3–7 days before
injection
INSTRUMENTATION
Camera: Large-field-of-view SPECT gamma camera
Dual-headed camera preferable
Collimator: Low-energy high-resolution collimator
Photopeaks: 20% window around 173 and 245 keV
Computer: 128 × 128 word mode matrix size
IMAGING PROCEDURE
4 hr: Planar images of abdomen and pelvis, 500,000
counts or 15 min; SPECT of abdomen
24 hr: Planar whole body imaging, 300,000 counts or 15
min; SPECT of abdomen and other regions as clinically
indicated
Oncology 281
Table 9-7 Radiation Dosimetry of In-111
Pentetreotide (OctreoScan)
Organ rads/mCi (cGy/37 MBq)
Large intestine 0.27
Kidney 1.8
Urinary bladder 1.02
Liver 0.4
Adrenal glands 0.25
Spleen 2.5
Thyroid 0.25
Testes 0.1
Red marrow 0.12
Total body 0.44
Target organ (highest radiation absorbed dose) appears in boldface type.
Image Interpretation
Normal uptake occurs in the thyroid gland, liver, gall-
bladder, spleen, kidneys, and bladder. The kidneys
retain considerable radiotracer and appear quite
intense even on delayed imaging (Fig. 9-18). Bowel
uptake should be absent or insignificant on the 4-hour
images. By 24 hours, significant bowel activity is often
seen normally.
Approximately 80–90% of tumors are visible by
4 hours. Due to decreasing background, more lesions
will be visible on 24-hour images. Many tumors can be
diagnosed with planar imaging (Figs. 9-19 and 9-20).
However, SPECT is essential in the abdomen. The region
between the kidneys can be difficult to scan because of
the high renal uptake.
Tc-99m NeoTect
NeoTect (Tc-99m Depreotide, Diatide) is a synthetic
peptide with high-affinity binding to somatostatin
receptors. Originally developed to diagnose neuroec-
todermal tumors, it proved inferior to octreotide, due
in part, to a high level of abdominal background activ-
ity. Delayed imaging at 24 hours was not realistic
because of the Tc-99m radiolabel. NeoTect binding
has been demonstrated in somatostatin receptor-posi-
tive pulmonary malignancies, and it has been
approved by the FDA for evaluating lung masses seen
on x-ray or CT.
The clinical indications are similar to that of F-18 FDG
(i.e., to evaluate a solitary pulmonary nodule for malig-
nancy). In one study, NeoTect differentiated malignancy
from benign processes with an overall sensitivity of 70%
and specificity of 86%. It was able to improve the predic-
tive value of malignancy from 85% with CT to 97% with
NeoTect. NeoTect appears to have a similar accuracy to
F-18 FDG PET.
Figure 9-18 Normal distribution of In-111 OctreoScan. A, Planar images at 4 hours show intense
uptake in the kidneys and spleen with liver activity nearly as intense. B, Delayed images at 24 hours
reveal diffuse bowel activity which is frequently seen and may obscure disease.

Figure 9-19 Metastatic carcinoid tumor on In-111 OctreoScan.Anterior planar views of chest
(A) and abdomen (B). Multiple sites of thoracic, mediastinal, and paratracheal uptake, as well as
uptake in the midabdomen, representing paraaortic adenopathy.The patient also had multiple other
sites not shown here, including many soft tissue metastases.
 Oncology 283

Figure 9-20 Anterior In-111 OctreoScan images reveal diffuse hepatic metastasis in a patient
with gastrinoma.

Method
A dose of 15–20 mCi (555–740 MBq) is administered
containing approximately 47 μg of Tc-99m radiolabeled
depreotide peptide. No patient preparation is neces-
sary, although patients should be well hydrated. Planar
and SPECT images of the chest are obtained between
2–4 hours after injection.

Dosimetry
Because NeoTect is a Tc-99m radiolabeled radiopharma-
ceutical, the radiation absorbed dose is low. The kidneys
are the critical organs with 0.33 rad/mCi (cGy/37 MBq).
The spleen receives 0.16 rad/mCi (cGy/37 MBq), and the
marrow receives 0.078 rad/mCi (cGy/37 MBq).

Figure 9-21 IgG antibody.The molecule can be digested

Tumor Therapy
In recent years, significant advances have occurred in the
use of targeted antibodies as a means to deliver radiation
therapy. Although preliminary results are promising, tar-
geted peptide radiotherapy is still investigational. It has
been shown that high doses of In-111 OctreoScan will
inhibit tumor growth in patients with somatostatin
receptor-positive tumors. In-111 emits an Auger electron
that travels only short distances but has been noted to
cause damage in the nucleus. This Auger electron dam-
age can be used for therapeutic purposes if higher doses
than the normal scan dose are given. High-energy beta-
emitting radioisotopes including Yttrium-90 (Y-90) and
Lutetium-177 (Lu-177) bound to octreotide have been
used effectively in early therapeutic trials against neu-
roendocrine tumors.
MONOCLONAL ANTIBODIES
Monoclonal antibody imaging has the potential for tar-
geting specific tumor types. In recent years, important
clinical advances have been made in the development of
antibodies for diagnosis and therapy. Four radiolabeled
monoclonal antibodies have been approved by the FDA
enzymatically by papain, resulting in three parts, two Fab′
fragments and one Fc fragment, or by pepsin to produce F(ab′)2
fragments and subfragments of Fc. Fab′ may be produced by
splitting the disulfide bond of F(ab′)2.
for imaging cancer of the colon, ovary, prostate, and lung.
Two radiolabeled antibodies have been approved for
therapy in B-cell lymphoma. Other antibodies are under
investigation and await the results of further clinical trials
and approval.
Background
Antibodies are proteins produced by lymphocyte plasma
cells in response to exposure to foreign antigens. An IgG
antibody consists of two identical heavy (H) and two
light (L) chains linked by a disulfide bridge (Fig. 9-21).
Each chain is made up of two regions. The variable
region (Fab′) is responsible specifically binding to a cell
surface antigen. The constant region (Fc) is involved
with cell destruction through complement fixation and
antibody-dependent cell cytotoxicity.
Each plasma cell produces one specific antibody
against a single antigenic determinant. However, animals
immunized with an antigen produce and secrete into
284 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-22 Monoclonal hybridoma antibody production.The process starts with injection of an
antigen into a mouse, causing proliferation of B-lymphocytes that can make antibody to the antigen.
The mouse spleen is removed and the B-cells are harvested. Many of the B-cells are capable of
making antibody to the specific antigen. If they were cultured at this point (left), they would make
a mix of antibodies and would soon die off. If instead the B-cells are mixed with mouse myeloma
cells in polyethylene glycol, some of the normal B-cells will fuse with the myeloma cells, producing a
population of hybridomas that can be cultured indefinitely.When this population is selectively
cloned for those that make the desired antibody, a pure culture of target antibody-producing cells
can be grown in great quantities. Its product is the desired monoclonal antibody.

their blood a mixture of antibodies from many plasma
cells, each against different antigenic determinants.
Some medically useful antibodies (e.g., gamma globulin)
have been produced in rabbits or other animals for
human use, but these polyclonal antibodies bind to multi-
ple different antigenic sites and are thus nonspecific.
Early antibody imaging studies used polyclonal antibod-
ies, often labeled with I-131. Although they showed
promising results in a variety of tumors, I-131 had imag-
ing and dosimetric disadvantages for diagnostic studies.
Kohler and Milstein won the Nobel Prize in 1975 for
describing a method for unlimited production of a single
monoclonal antibody (MoAb). Myeloma cancer cells
from a mouse are fused with lymphocytes from the
spleen of mice immunized with a particular antigen (Fig.
9-22). These “hybridoma” cells retain both the specific
antibody production capacity of the lymphocytes and
the immortality of the myeloma cancer cells.
Immunoassays screen the hybrid cells to identify the
specific cell line that produces the murine (or mouse)
MoAb clone desired. A MoAb with high affinity and
specificity for the antigen of interest is harvested. The
individual hybridoma cells can be maintained in culture
to produce large quantities of the monoclonal antibody
for future use.
MoAbs that are produced by the immunization of
mice are mouse proteins, and as such are recognized by
the human immune system as foreign. The human
immune system mounts an immunological response
against the MoAb. This human antimouse antibody
(HAMA) response ranges from mild with fever and hives
to severe with shortness of breath and hypotension. The
reaction may even be fatal as a result of anaphylaxis.
The HAMA response is related to the amount of anti-
body the patient is exposed to as well as the size of the
antibody. Rather than delivering a whole intact antibody,
a fragment can be used which will be less immunogenic.
The active regions are kept and the large constant por-
tions of the antibody that contribute the most to the
HAMA response are deleted.
Proteolytic enzymes such as pepsin and papain will
cleave off the Fc portion of the antibody. Pepsin produces

a larger F(ab′)2 fragment and papain a smaller F(ab′)
fragment (Fig. 9-21). These fragment antibodies not
only have fewer immunogenic side effects but are also
better suited for imaging. These fragments are cleared
from the background much more rapidly, thus improving
tumor detection.
Despite the development of MoAb fragments, the
potential for serious reactions remains a serious clinical
concern. In an attempt to overcome the problems with
using foreign mouse proteins, new production methods
have been developed. One first step was the creation of
chimera MoAbs by replacing the murine Fc portion of
the antibody with a human component. Further steps
have lead to fully human MoAbs through phage display
and recombinant DNA technology. Although not yet
available with the FDA-approved MoAb imaging agents
described here, these new antibody production tech-
niques will likely play an important clinical role in the
near future.
The MoAb imaging agents currently available for clini-
cal use are derived from both whole murine antibodies
and fragments. They are selected based on a high level
of tumor-specific binding. Many tumors express specific
antigens on their surfaces, which can be targeted. An
example of this is carcinoembryonic antigen (CEA) on
colon cancer cells. Other tumors express increased
numbers of normal antigens or receptors on their sur-
faces, such as the CD20 receptor on certain lymphomas.
Radiolabeling
Chemists have attached various radionuclides (such as
I-131, I-123, In-111, and Tc-99m) to MoAbs. Each has
distinct advantages and disadvantages (Table 9-8).
Radiolabeling must be done without changing the anti-
body’s immunoreactivity or biological properties so
that the resulting radiopharmaceutical can be used suc-
cessfully for immunoscintigraphy.
Although I-131 can be easily bound to other molecules,
it creates very limited images because of its physical prop-
Table 9-8 Radionuclides Used for Immunoscintigraphy: Advantages and Disadvantages
Radionuclide Energy (keV) Half-Life Advantages
Technetium-99m 140 6 hr Pure gamma, inexpensive,
high-photon flux
Indium-111 173, 247 2.8 days Gamma emitter
Iodine-123 159 2.8 days Gamma emitter, ease of labeling
Iodine-131 364 8 days Ease of labeling
Oncology 285
erties. I-123 has very good imaging characteristics but has
been difficult and expensive to obtain in the past, and it is
only recently that production methods became available
that create a product with lower dosimetry by eliminating
I-124 and I-125 radiocontaminants.
In-111 was frequently used for MoAb radiolabeling
because it has better imaging characteristics than I-131.
The 2.8-day half-life of In-111 allows time for the slow
radiopharmaceutical accumulation and background clear-
ance of whole antibodies. The best target-to-background
ratio for imaging occurred 48–72 hours after injection.
With the development of antibody fragments and
their more rapid background clearance, high target-to-
background ratios can be obtained on the day of injec-
tion. Thus, imaging with Tc-99m–labeled MoAbs became
a possibility. The optimal imaging characteristics of
a technetium label may lead to significant improvements
in lesion detection.
Clinical Imaging Applications
The FDA has approved four radiolabeled monoclonal
antibodies for oncological diagnostic imaging: OncoScint
for colorectal and ovarian cancer, CEA-Scan for colorectal
cancer, ProstaScint for prostate cancer, and Verluma for
small cell carcinoma of the lung.
Colorectal Cancer
Colorectal cancer is the second most common malig-
nancy in the United States. Prognosis is strongly
dependent on tumor staging (Box 9-10). Five-year sur-
vival is 85% with localized disease, 50% with regional
spread, and less than 7% with distant metastases. The
first recurrence occurs at a single site in 75% of cases.
These sites include the liver (33%), local or regional sites
(21%), intra-abdominal sites (18%), and retroperitoneal
lymph nodes (10%).
CEA arises from ectodermally-derived epithelial cells
of the digestive system. It is normally only expressed
during embryological development but is also found in
Disadvantages
Complex chemistry, short half-life,
high counts in kidney and bladder
Affinity for liver and RES, delayed
imaging possible
Dehalogenates, cyclotron produced,
expensive due to short half-life
Dehalogenates, low count rate, poor
image quality, high radiation dose
286 NUCLEAR MEDICINE: THE REQUISITES
Box 9-10 Dukes’ Classification for
Colorectal Cancer Staging
DUKES A Involvement into bowel wall but not
beyond muscularis
DUKES B Spread beyond pericolonic tissue;
No lymph nodes involved
DUKES C Locoregional lymph node involvement
DUKES D Distant metastasis
colorectal cancer and other solid tumors. Over 95% of
colorectal cancers express membrane surface CEA. It is
shed into the bloodstream and is detectable in 65% of
patients with colorectal carcinoma. Although serum
CEA levels are used as a marker of residual and recurrent
tumor, nearly one-third of patients with recurrence do
not have elevated serum CEA levels.
Most recurrences and metastases occur in the
abdomen and pelvis. Colonoscopy and barium studies
produce a low yield in determining sites other than local
recurrence because they detect only intraluminal dis-
ease. CT is a sensitive method for detecting metastatic
colon cancer. However, CT often misses lymph node
involvement, especially as lymph nodes must be enlarged
(>1 cm) before they are usually considered abnormal. In
addition, postoperative and radiation changes may
obscure tumor. F-18 FDG-PET has proven highly effec-
tive in detecting distant colorectal metastasis. However,
it has limitations. Local lymph node involvement is fre-
quently missed, due in part to adjacent bowel activity.
Also, the sensitivity of F-18 FDG PET is lower in certain
tumors types (e.g., mucinous rectal tumors).
Radiopharmaceuticals
OncoScint CR/OV OncoScint (satumomab) was
the first monoclonal antibody approved as a tumor-
imaging agent by the FDA (1994). It is a B72.3 murine
IgG monoclonal antibody directed against a high
molecular-weight tumor-associated glycoprotein (TAG-
72), which is expressed by the majority of colorectal and
ovarian cancers. Although not currently available
commercially in the United States, OncoScint detects
several types of tumor, most commonly ovarian and
colorectal cancer. In ovarian carcinoma, sensitivity and
specificity were 95% and 56%, respectively. In colorectal
carcinoma, it showed sensitivities ranging from 70–89%
and a specificity of 76%.
The antibody is linked to In-111 by conjugation to the
Fc portion, which avoids the area of active antigen bind-
ing and preserves the immunoreactivity of the antibody
(Fig. 9-23). OncoScint is approved for localization and
determination of the extent of extrahepatic metastatic
tumor in patients with known colorectal or ovarian can-
Antibody Linker Substance
Carbohydrate attachment
Chemically stable Gamma-emitting
chelator radionuclide
CYT-356 GYK-DTPA In-111
B72.3
Figure 9-23 Indium-111 OncoScint CR/OV and In-111
ProstaScint formulation.The site of attachment of the linker does
not interfere with the effector or binding functions of the
antibody. (Courtesy of Cytogen, Princeton, NJ.)
cer (Fig. 9-24). Although the urinary clearance is lower
than fragment antibodies, leading to low levels of kidney
uptake, there is significant liver background which
interferes with the detection of liver metastasis.
Tc-99m CEA-Scan CEA SCAN was approved in 1996
for imaging of colorectal cancer. It is a Tc-99m-labeled Fab′
fragment of the CEA antibody IMMU-4. Removal of the
Fc group of IgG, the most immunogenic part of the
molecule, eliminates much of the immunogenicity
ordinarily observed with mouse-derived antibody products.
Pharmacokinetics The pharmacokinetics of Onco-
Scint and Tc-99m CEA-Scan are very different, largely
because the former radiopharmaceutical is a whole
antibody and the latter is an antibody fragment. Table 9-9
compares the pharmacokinetics of OncoScint and CEA-
SCAN. Liver metabolism is low but renal activity is high
compared with that for the whole antibody (Fig. 9-25).
An advantage of the Tc-99m-labeled Fab′ fragment is
its rapid renal clearance from the blood, allowing for
same day high tumor-to-background ratio imaging. At 1,
5, and 24 hours after infusion, 63%, 23%, and 7%, respec-
tively, of the injected dose is present in the circulation.
Over 24 hours, 28% of the dose is excreted in the urine.
Indications The FDA approved CEA-Scan for
detection and localization of recurrent, metastatic, and
occult colorectal carcinoma in patients with his-
tologically confirmed colorectal carcinoma (Fig.9-26). Its
major role to date has been in the evaluation of recurrent
disease. The two clinical indications are a patient with a
rising serum CEA level but negative conventional
imaging, and a patient with known potentially resectable
disease who requires preoperative evaluation to exclude
the presence of unresectable disease. The CEA-Scan can
assure the surgeon that the patient has no other
metastatic disease that would contraindicate surgical
treatment. Due to the increasing clinical role of PET,
 Oncology 287

Figure 9-24 Distant recurrent colorectal cancer with indium-111 OncoScint.Tumor uptake is
seen in the left supraclavicular nodes and left hilum (A), in the periaortic nodes, and more diffusely
throughout the abdomen (B and C). Retrosternal uptake was detected with SPECT.

Table 9-9 Comparison of OncoScint CR/OV and

CEA-SCAN

OncoScint
CR/OV CEA-SCAN

Radionuclide In-111 Tc-99m

Monoclonal B72.3, whole IMMU-4 reactive
antibody type antibody with CEA, Fab′
fragment
HAMA 40% <1%
Liver metabolism High Low
and uptake
Renal metabolism Low High
and uptake
Plasma half-life Slow (50 hr) Rapid (initial T 12⁄ 1 hr,
final T 12⁄ 13 hr)
Urinary excretion 10% at 72 hr 28% at 24 hr

Figure 9-25 Technetium-99m CEA-SCAN normal distribution.

CEA-Scan is often reserved for F-18 FDG PET negative
tumors (e.g., mucinous cellularity) or for cases where
PET is unavailable.
The role of Tc-99m CEA in primary disease is not well
established. Possible applications might be the detec-
tion of synchronous lesions, preoperative determination
of the extent of regional disease, or search for occult
metastases.
Accuracy In a multicenter trial, 192 patients with
colorectal carcinoma were imaged with OncoScint
CR/OV and CT. The overall sensitivity was 69%,
specificity was 76%, positive predictive value was 97%,
and negative predictive value was 19%. Scans detected
occult disease in 10% and changed patient management
in 25%. Although CT was more sensitive than antibody
imaging of the liver, OncoScint was superior for the
pelvis and extrahepatic abdomen (Table 9-10). The
The kidneys have the greatest uptake of the radiopharmaceutical.
Renal clearance into the bladder is seen. Cardiac and vascular
blood pool is prominent. Lesser distribution is seen in the liver and
spleen.The focal uptake above the bladder is the uterine blood
pool.
combined sensitivity of CT and OncoScint immuno-
scintigraphy (88%) was higher than the sensitivity of
either study alone.
In a multicenter trial of 210 patients with advanced recur-
rent or metastatic colorectal carcinomas,the sensitivity of Tc-
99m CEA for detection of metastatic colon cancer in the
abdomen, liver, and pelvis was superior to CT (Table 9-10).
Tc-99m CEA was superior to CT in the extrahepatic
abdomen and pelvis. The accuracy of CT and CEA-Scan was
similar in the liver. The combination of Tc-99m CEA and
CT increased the overall sensitivity from 66% to 78% while
only slightly decreasing specificity (from 89% to 83%).
Tc-99m CEA is superior to OncoScint in several
respects. First, it has better imaging characteristics
288 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-26 Technetium-99m CEA, local recurrence of colorectal cancer. Patient had rising
serum CEA level several months after primary resection of tumor in the rectosigmoid area.This
reconstructed volume display of sequential projection angles (Ang = degrees) shows tumor
recurrence in the rectal area (arrowheads).

because of the Tc-99m radiolabel. Second, the antibody
fragments of Tc-99m CEA clear more rapidly. This results
in a higher target-to-background ratio at an earlier imaging
time (day 1 versus day 2 or 3). Because of the absence of
high liver uptake with Tc-99m CEA-Scan, it is also superior
to In-111 OncoScint in detecting liver metastases.
Although liver metastases are often photopenic with
OncoScint, they are usually hot or target lesions with
Tc-99m CEA. Finally,CEA-Scan has a much lower incidence
of HAMA response, less than 1% versus 40% for OncoScint.
One disadvantage of the CEA F(ab′) fragment is the high
level of renal uptake,which can obscure disease.
Methodology Imaging protocol for Tc-99m CEA
Scan is described in Box 9-11.
Dosimetry The estimated radiation absorbed patient
doses for OncoScint CR/OV and CEA-Scan are detailed in
Table 9-11. The highest radiation dose from In-111
OncoScint occurs in the spleen and red marrow. For
Tc-99m CEA the highest dose is in the kidney, followed
by the urinary bladder and spleen.
 Oncology 289

affected are available. HAMA can alter the biodistribu-

Table 9-10 Imaging Sensitivity of OncoScint and tion and pharmacokinetics of MoAbs and may interfere
CEA-SCAN versus Computed with the quality or sensitivity of the imaging study.
Tomography by Anatomical Site
Ovarian Cancer
Sensitivity for colorectal tumor localization (%) Ovarian cancer is the fourth most frequent cause of can-
Anatomical In-111 Tc-99m
cer deaths in women. The overall 5-year survival rate is
site OncoScint CT CEA-Scan CT 39%. Ovarian cancer is difficult to diagnose and stage
with current imaging methods because it frequently
Pelvis 74 57 69 39 metastasizes as small (<2 cm) miliary peritoneal implants
Abdomen 66 34 55 32
(extrahepatic)
not detectable on CT. Nor can CT detect tumor in nor-
Liver 41 84 63 64 mal-sized lymph nodes or distinguish adhesions or scar
from tumor. Although F-18 FDG PET has added sensitiv-
ity when used with the CT, the pattern of uptake may be
misleading due to nonspecific normal bowel activity.
Box 9-11 Tc-99m CEA-Scan Sample Serum CA-125 assay, a tumor marker, has a high false-
Protocol negative rate and does not predict the location or extent
of disease. Exploratory laparotomy is the best approach
to surgical staging. However, it does not detect extraab-
PREPARATION dominal tumors, is expensive, has a 20% complication
None rate, and gives false-negative results in 20–50% of patients
based on the results of second-look surgery.
RADIOPHARMACEUTICAL
In a multicenter trial of patients with primary or
Dose: 30 mCi (1110 MBq) intravenously
recurrent disease, OncoScint had a sensitivity of 60–70%
INSTRUMENTATION and a specificity of 55–60% for ovarian cancer. It was
Camera: Large-field-of-view gamma camera, Dual-headed superior to CT for patients with recurrent disease and
camera preferable carcinomatosis (60% versus 30%). Tc-99m CEA-Scan has
Collimator: Low-energy, high-resolution not been used clinically or approved for ovarian cancer.
Photopeaks: 15% symmetric window around 140 keV F-18 FDG PET has proven useful in ovarian carcinoma,
Computer: 128 × 128 word matrix size although it may miss very small tumors which often stud
the peritoneum.
IMAGING PROCEDURE
Image: imaging 2 hr after injection Prostate Cancer
Planar images: 10 min/view spot images chest to pelvis Cancer of the prostate is the most frequently diagnosed
SPECT:Abdomen and pelvis, with tow-headed camera:
malignant tumor in men in the United States and the sec-
60 stops/head, 40 sec each
ond leading cause of cancer death. Its incidence is
Optional 24-hr planar imaging (20 min/view) or SPECT
(50% increased acquisition time) increasing. The 5-year survival is approximately 50%.
Although many patients have symptoms for which they
seek medical evaluation, the diagnosis is often suspected
on the basis of screening prostate-specific antigen (PSA)
Adverse Effects The incidence of side effects with levels drawn on men older than 50 years. Ultrasound-
OncoScint is less than 4%. Most are not serious and are guided needle biopsy is used to obtain tissue from suspect
readily reversible, generally without intervention. Adverse nodules.
effects with Tc-99m CEA have also been uncommon and Staging of prostate cancer is based on the combina-
self-limiting. tion of physical examination, histopathological Gleason’s
The incidence of elevated HAMA with CEA-Scan is score, and serum PSA. Bone scans are indicated for
less than 1%, compared with a 40% incidence with patients with serum PSA greater than 10–20 ng/ml or
OncoScint. Generally HAMA levels decrease with time with a high Gleason’s score. In patients who have under-
and half of cases become seronegative. This has implica- gone prostatectomy, any increase in PSA is suspicious.
tions for using OncoScint in a serial manner to evaluate Lymph node involvement is the most common pattern of
the effectiveness of therapy or as a prelude to therapy metastatic spread, usually occurring in a stepwise fashion
with a MoAb. At present, only a single administration has from periprostatic or obturator nodes, to internal and
been approved. HAMA can interfere with murine-based external iliac nodes, and then to common iliac and
immunoassays of CEA and CA-125, producing falsely high periaortic nodes (Fig. 9-27). Frequent sites of distant
values. Alternative assay methods that are not adversely metastases are the skeleton, liver, and lungs.
290 NUCLEAR MEDICINE: THE REQUISITES

Table 9-11 Dosimetry of Approved Monoclonal Imaging Agents

In-111 Satumomab Tc-99m Arcitumomab In-111 Capromab

Pendetide OncoScint CEA-Scan Pendetide ProstaScint
rads/5 mCi rads/30 mCi rads/5 mCi
Organ cGy/185 MBq cGy/1110 MBq cGy/185 MBq

Gallbladder wall 7.3

Large intestine 3.1 7.6
Kidney 9.7 11.1 12.4
Urinary bladder 2.8 1.8 2.2
Liver 15.0 1.1 18.5
Lungs 4.9 5.6
Adrenal glands 4.5 5.32
Ovaries 2.9 0.5
Heart wall 3.2 7.8
Spleen 16.0 1.8 16.3
Thyroid 1.5 1.4
Testes 1.4 0.5 5.6
Red marrow 12.0 4.3
Total body 2.7 0.5 2.7

Target organ (highest radiation absorbed dose) appears in boldface type.

ease at surgery. The rate of local recurrence after sur-

Figure 9-27 Pelvic and abdominal lymph node anatomy.
Knowledge of this anatomy is critical for proper interpretation of
indium-111 ProstaScint studies.
Initial therapy involves either surgery or radiation
therapy. Radical prostatectomy, the best chance for cure,
is not undertaken when there is evidence of nodal
involvement or distant spread. Thus, assessing the status
of pelvic lymph nodes draining the prostate gland is criti-
cal to staging and management. CT and MRI have lim-
ited value owing to their low sensitivity for detecting
nodal involvement. F-18 FDG PET has low sensitivity for
prostate cancer. Even when inactions are favorable that
the tumor is contained intracapsularly before surgery,
patients are frequently found to have extracapsular dis-
gery is 15–20%.
Lymphadenectomy, the most accurate technique for
detecting nodal involvement, may fail and lead to surgery
for patients with occult disseminated disease. Patients
with high PSA levels and a high Gleason’s score are usu-
ally treated with local radiation therapy because they are
at risk for local recurrence. Radiation therapy can be
performed as the initial treatment or following radical
prostatectomy.
If after initial treatment the PSA fails to fall to unde-
tectable levels or subsequently rises, residual or recurrent
cancer is likely. Radiation therapy of the prostate fossa or
the pelvis is often given, even in the absence of positive
biopsy or positive imaging results. If disease is localized
to the prostate fossa or pelvis, radiation therapy offers the
potential for effective treatment. However, if recurrence
involves periaortic lymph nodes or other distant sites,
radiation therapy exposes the patient to significant mor-
bidity with no potential for cure,owing to the presence of
tumor outside the radiation therapy field. In this situa-
tion, In-111 ProstaScint can play an important role.
Indium-111 ProstaScint
In-111 ProstaScint (Capromab Pendetide, Cytogen,
Princeton, NJ) is a conjugate of the monoclonal antibody
7E11-C5.3 (CYT-356), a linker-chelator (GYK-DTA), and
In 111 (Fig. 9-28). This is an intact murine immunoglo-
bulin reactive with prostate-specific membrane antigen
(PMSA), a glycoprotein expressed by more than 95% of
prostate adenocarcinomas. ProstaScint was approved
in 1996 as an imaging agent for the detection of soft tissue
metastases for patients with prostate cancer who were at
high risk for metastatic disease.

Figure 9-28 Indium-111 ProstaScint shows paraaortic and
mesenteric lymph nodes.This planar abdominal image showed no
change in distribution between days 3 and 6, excluding bowel
activity as the cause for this activity.
Pharmacokinetics and Normal Distribution In-
111 ProstaScint follows a monoexponential clearance
pattern with a biological half-life of 72 hours. Ten
percent is excreted in the urine within 72 hours, and
a smaller amount is excreted through the bowel.
Normal distribution includes the liver, spleen, bone
marrow, and blood pool structures. Clearance occurs
into the bowel and bladder.
Accuracy In a multicenter trial, 152 patients with
a tissue diagnosis of prostate cancer scheduled for pelvic
lymphadenectomy had ProstaScint scans. Other
standard noninvasive imaging including bone scans, CT,
and MRI were negative or equivocal. The patients were
considered at high risk for the presence of lymph node
metastases based on PSA or Gleason’s score. The
imaging results were correlated with histological analysis
of pelvic lymph nodes. ProstaScint correctly identified
lymph node metastases in 40 of 64 patients (sensitivity
62%), compared with a sensitivity of 4% for CT and 15%
for MRI. Of 88 patients without pelvic nodal metastases,
63 were correctly identified as normal (specificity 72%)
(Table 9-12). The specificity of ProstaScint may actually
be higher than these results suggest because 15 patients
with a false-positive study had biochemical evidence of
disease after radical prostatectomy, suggesting that
disease was missed.
Results were similar in a multicenter series of 183
patients in whom residual or recurrent prostate cancer
after radical prostatectomy was strongly suspected based
on rising PSA levels, but bone scans and standard imaging
methods gave negative results. Although the accuracy of
Oncology 291
Table 9-12 Comparison of ProstaScint with Pelvic
Lymph Node Dissection
ProstaScint ProstaScint
scan scan
positive negative
Biopsy 40 24 Sensitivity 62%
positive
Biopsy 24 63 Specificity 72%
negative
ProstaScint scanning is only fair, it far surpasses all other
available imaging modalities.
Minor adverse events have been reported in 4% of
patients. Most common have been liver enzyme eleva-
tions, hypotension, and hypertension, each occurring
in 1% of patients or less. Elevated HAMA titers have
been observed in 8%. A similar incidence (4%) of
adverse events was seen in patients undergoing
repeated injections.
Indications In-111 ProstaScint is indicated for
the assessment of nodal metastases in two different
situations. First, it can be used to stage patients at high
risk for pelvic lymph node metastases but with
apparently localized disease after standard diagnostic
evaluation tests. Most commonly, In-111 ProstaScint is
used in postprostatectomy patients when occult
metastatic disease is suspected because of a rising PSA
level but the standard workup is negative or equivocal.
Radiation therapy is indicated if disease is localized to
the prostate bed and pelvic lymph nodes, but not if the
scan shows activity in periaortic lymph nodes or other
distant sites. In the latter case hormonal manipulations,
systemic chemotherapy, or orchiectomy would be more
appropriate treatment options.
Methodology An imaging protocol for In-111
ProstaScint is described in Box 9-12. SPECT of the
abdomen and pelvis is mandatory. Blood pool images are
necessary for correct interpretation. They may be
acquired either by imaging on day 1 at 30 minutes after
In-111 ProstaScint injection or,preferably,by radiolabeling
the patient’s red blood cells and acquiring dual-isotope
Tc-99m RBC and In-111 ProstaScint planar and SPECT
images at 96 hours (3–5 days). The day 3 images often
have problematic bowel activity, and the day 5 images are
occasionally limited by a low count rate. Review of the
two study days together gives us the most confidence in
interpretation. In addition, software fusion packages are
available which can combine the ProstaScint images with
a CT to optimize lesion localization or characterization. If
available, SPECT-CT can be very helpful.
Dosimetry The highest In-111 ProstaScint radiation
dose is received by the liver, followed by the spleen
292 NUCLEAR MEDICINE: THE REQUISITES

BOX 9-12 Indium-111 Capromab Pendetide ProstaScint Imaging: Dual Isotope Protocol
Summary

RADIOPHARMACEUTICAL CAPROMAB PENDETIDE

Dose: 5 mCi (185 MBq) In-111 ProstaScint intravenously over 5 minutes. Observe patient 30 minutes

PATIENT PREPARATION
Well hydrated, bowel cleansing preparation 2 days before imaging

PLANAR IN-111 PROSTASCINT IMAGING

96 Hours after injection: patient voids
Whole body images performed by either multiple overlapping spot views (10 min/view, 256 × 256 matrix) OR whole body
scan mid femur to head (4-6 cm/min)
Collimator: Medium-energy collimator
Photopeaks: 20% window around 173, 247 keV
Computer: 256 × 256 matrix size for spots, 512 × 1024 for whole body scan
Draw and label red blood cells

SPECT IMAGING PROCEDURE

96 hours (3-5 days) Image after planar images:
patient voids
Re-administer 2-3 mCi (74-111MBq) Tc-99m tagged RBCs
Positioning: SPECT pelvis make sure symphysis included

ACQUISITION
128 × 128 matrix, 64 stops/head; 50-60 sec/stop
140 keV peak window at 5%; 173 keV window at 15%; 247 keV window at 20%

PROCESSING
Fusion software for overlie and localization with CT
Occasionally, repeat delayed imaging required to permit time for blood pool, bladder, or bowel clearance

and kidneys. Table 9-11 lists dosimetry estimates for
the body.
Interpretation There is a steep learning curve for
interpretation of In-111 ProstaScint SPECT studies.
The FDA approved this radiopharmaceutical for clinical
use and interpretation only by physicians who have
undergone specific training in the acquisition and
interpretation of these studies. There are several
reasons for the concern about interpretive difficulty. In the
pelvic SPECT, there is a paucity of normal anatomical
landmarks. The individual cross-sectional images have low
counts and poor resolution. Bowel and bladder clearance
can complicate interpretation. Increased uptake may be
seen in the prostate bed following radiation therapy.
The dual-isotope acquisition method allows single-day
imaging and perfect image registration of the two stud-
ies. For correct interpretation, the physician must be
familiar with pelvic lymph node anatomy and common
patterns of tumor spread (Fig. 9-28). The lymph nodes
lie along the blood vessels. Therefore, the In-111
ProstaScint images must be carefully correlated with the
blood pool images,because right-to-left vascular asymme-
tries may otherwise be misinterpreted as nodal disease
on the In-111 ProstaScint images. An abnormal scan
shows increased uptake in the prostate fossa, at pelvic,
abdominal, or chest lymph node sites, or less commonly
in bony structures. Pelvic lymph node metastases are
best seen on the SPECT studies and rarely seen on planar
studies (Fig. 9-29). However, both planar and SPECT
images may show periaortic lymph or thoracic lymph
node metastases. ProstaScint is considerably less sensi-
tive (50%) than bone scan for detecting bone metastases.
SPECT images software fused with CT or SPECT-CT
improves anatomical certainty in interpretation.
Lung Carcinoma
Lung cancer can be divided into two distinct diseases
based on tumor biology and chemotherapy responsive-
ness: small-cell carcinoma of the lung (SCLC) and
nonsmall-cell lung cancer (NSCLC). SCLC accounts for
25% of all new lung cancers in the United States. Survival
is poor, 18% at 5 years with limited disease and only 2%
with distant metastases. Two thirds of patients with SCLC
have metastatic spread at the time of diagnosis, thus only
one third would be expected to respond to local therapy.
Staging determines the extent of disease at the time of

Figure 9-29 Indium-111 ProstaScint detects metastases to regional nodes. Sequential coronal
images show prominent uptake in external iliac nodes (arrowhead).
presentation and guides therapy. Patients with limited dis-
ease are treated with local radiation therapy and systemic
chemotherapy, whereas patients with extensive disease
receive palliative treatment with chemotherapy alone.
NSCLC is primarily a surgical disease; resection is the
treatment of choice for localized disease. Accurate staging
is essential to determine whether the patient is potentially
curable. The standard diagnostic imaging staging method
is CT. Mediastinoscopy with lymph-node biopsy is indi-
cated to evaluate enlarged or equivocal lymph nodes. The
patient is considered a candidate for primary tumor resec-
tion if no evidence of tumor spread to extrathoracic sites,
the contralateral chest, or the mediastinum is found. CT
relies on lymph-node size to detect metastatic disease.
However, normal-sized nodes may contain microscopic
tumor. Because of this, CT lacks sensitivity and may under-
estimate the extent of lung cancer. Thus patients may
undergo unnecessary surgery. In addition, enlarged nodes
may be due to reactive hyperplasia or infection,resulting in
false-positive findings. Although mediastinoscopy may
improve accuracy,it is invasive and expensive.
Tc-99m Verluma
Verluma (Tc-99m Nofetumomab,DuPont Pharmaceuticals)
is a Tc-99m-labeled Fab′ fragment of a murine IgG2b mono-
clonal antibody NR-LU-10 directed against a 40-kilodalton
glycoprotein expressed on a variety of carcinomas, includ-
ing SCLC,NSCLC,and cancers of the breast,ovary,prostate,
colon, and rectum. Tc-99m Verluma was approved by the
FDA in 1996 as a diagnostic imaging agent for staging of
patients with newly diagnosed SCLC.
Pharmacokinetics Renal clearance is the main route
of excretion,with 64% of the injected dose excreted within
the first 22 hours. The secondary route of elimination is
hepatobiliary, with clearance into the gallbladder and
intestines. HAMA develops in only 6% of patients.
Accuracy Tc-99m Verluma was compared with
conventional diagnostic methods in a multicenter trial of
Oncology 293
96 patients with SCLC, of whom 42% had limited and
58% had extensive disease as evaluated with standard
imaging modalities. Tc-99m Verluma correctly staged
82% of patients. The positive predictive value for
demonstrating extensive disease was 94%. Sensitivity for
tumor detection was 77%, compared with 88% for
a battery of standard diagnostic tests. Tc-99m Verluma
had the highest accuracy for clinical staging of any single
diagnostic test.
Although approved for SCLC,Tc-99m Verluma is taken
up by other tumors. In addition to NSCLC, uptake has
been reported in gastrointestinal, breast, ovarian, pancre-
atic, renal, and cervical cancers. The ultimate role of
Tc-99m Verluma in these cancers is uncertain and
will require further investigation. However, Tc-99m
Veraluma is not commercially available.
Methodology Imaging is performed about 18 hours
after injection of the radiopharmaceutical (30 mCi or
1110 MBq). SPECT of the chest is routine.
Monoclonal Antibody Therapy
of Lymphoma
Non-Hodgkin’s lymphoma (NHL) is the most common
hematologic cancer and the sixth most common cause of
death. The most common forms of NHL are follicular
cell and diffuse large B-cell lymphoma. Although large
B-cell NHL can be cured, low-grade follicular NHL is con-
sidered incurable. Initial response rates are high but
patients ultimately relapse and die. The mean survival is
8–10 years with conventional chemotherapy and radia-
tion therapy. Response rates of recurrent episodes
become progressively less, and these patients frequently
transform into a higher-grade lymphoma.
There are several immunologic targets on the B-cell
surface that MoAbs can specifically bind. The CD20 anti-
gen is present on 90% of the B-cell lymphomas on mature
294 NUCLEAR MEDICINE: THE REQUISITES

Table 9-13 Comparison of Low-Grade NHL

Monoclonal Antibody Therapies: Y-90
Zevalin and I-131 Bexxar

Y-90 Zevalin I-131 Bexxar

Radiolabel half-life 64 hrs 8 days

Beta particle 2.293 MeV 0.606 MeV
5 mm path 8 mm path
Gamma emission No Yes; 364 keV
Pretreatment dosimetry No Yes
Pretreatment unlabeled Rituximab Tositumomab
antibody chimeric murine
HAMA 1-2% 60%
Outpatient therapy + +/−

B-lymphocytes. CD20 is not expressed on the mem-

branes of normal hematopoietic cells,antibody-producing
mature plasma cells, early B-cell precursors, or other lym-
phoid tissues.
This CD20 antigen is the target for the currently
approved immunologic therapies Yttrium-90 ibritu-
momab (Y-90 Zevalin) and I-131 tositumomab (I-131
Bexxar) (Table 9-13). Radiolabeled MoAb therapies show
better tumor response than treatment with nonradiola-
beled MoAb. They each recruit immune system response
to aid in tumor cell killing, but they also irradiate the cell
they are bound to and cause “crossfire” killing of tumor
cells. Crossfire is the result of beta-radiation traveling
a short distance from the site of antibody binding and
killing adjacent tumor cells not bound to the MoAb.

Y-90 Ibritumomab tiuxetan (Y-90 Zevalin)
Y-90 Ibritumomab tiuxetan (Zevalin, IDEC-Y2B8) was the
first FDA-approved radiolabeled antibody therapy agent
(Fig. 9-30). It is a murine IgG1 kappa monoclonal anti-
body that targets the CD20 receptor on lymphocytes.
The short half life of the radiotracer is ideal for effective
clearance. Tiuxetan is a chelator that binds either In-111
or Y-90 and provides a stable link between the radioiso-
tope and the antibody.
Yttrium-90 has a physical half-life of 64 hours and is
a high energy (2.29 MeV) pure beta-emitter. The beta-
particle travels only a short distance of 5 mm. An effec-
tive dose of radiation will be deposited very close to the
site of radiolabeled antibody binding. No imaging can be
done with this agent because no significant radiation
leaves the patient (e.g., no gamma ray emission).
However, this also means the dose requires no special
shielding and the therapy can be done as an outpatient.
In fact, few special radiation safety precautions are
needed. This situation contrasts with I-131 Bexxar ther-
apy, which is stored and administered through heavily
shielded equipment.
Figure 9-30 Biodistribution scan for Y-90 zevalin lymphoma
therapy. Prior to therapeutic infusion of the pure beta-emitter,Y-90
zevalin, a biodistribution scan is performed with In-111
ibritumomab tiuxetan (zevalin) planar images. In-111 Zevalin
images at 24 hours (A) and 48 hours (B) show an expected
biodistribution. Normal activity is seen in the kidneys, liver and
blood pool. Expected uptake is present in the lymph nodes
involved with tumor in the midabdomen, left groin and left pelvis.
This patient may go on to receive the therapeutic Y-90 zevalin
dose.
Indications
Y-90 Zevalin is used for the treatment of relapsed, refrac-
tory, or transformed CD20+ non-Hodgkin’s lymphoma. It
is absolutely contraindicated in patients with a known
hypersensitivity reaction to murine proteins (HAMA),
greater than 25% tumor involvement of marrow, and
those with impaired marrow reserves. Patients should
not have had myelotoxic therapies with autologous bone
marrow transplant or stem cell rescue. External beam
radiation should not have involved over 25% of the mar-
row. The neutrophil count must be over 1500 cells/mm3
and platelet count must be over 100,000.
 Oncology 295

Dosimetry heart, urinary tract, or bowel uptake. Any of these

Y-90 Zevalin has a biologic half-life of 46–48 hours with
a typical dose to the tumor of 15–17 Gy. Elimination is
primarily through the urine, although most of the agent
remains in the body. The radiation dose to various
organs is listed in Table 9-14.
Methods
The protocol consists of three parts (Table 9-15).
Rituximab (Rituxan) is given to block CD20 antigens on
cells circulating in the blood and spleen. The patient
must be closely monitored during this infusion as serious,
potentially fatal reactions can occur. Then, within 4 hours
of receiving the rituximab, 5.0 mCi (185 MBq) of In-111
labeled zevalin is administered over 10 minutes. Then
whole body planar images are done within 2–24 hours
and between 48–72 hours later to assess distribution of
the radiopharmaceutical. An optional third image can be
obtained over the next 90–120 hours. Normally, low uri-
nary tract uptake and low bowel activity is seen, but fairly
high liver and spleen uptake are present. The blood pool
activity should markedly decrease over the studies.
Altered blood pool distribution would include activity
increasing rather than decreasing over time in lung, liver,
changes could lead to unacceptable radiation to the
organ in question, such as the kidneys.
Toxicity
Significant side effects may occur from this therapy.
Usually within 7–9 weeks, blood counts reach a nadir
(median neutrophil count 800, platelet count 40,000,
hemoglobulin 10.3). The cytopenia may last from 7
to 35 days. In particular, roughly 7% of neutropenic
patients are prone to febrile neutropenia and infections.
Thrombocytopenia may result in hemorrhage. Up to
one third of patients will subsequently transform to
a more aggressive lymphoma. It is uncertain if this is a
side effect of therapy or the natural course of the disease.
A small number (1.4%) of patients will develop myelodys-
plasia or acute myelogenous leukemia.
Results
This is a very effective therapy. Overall, 67–83% of
patients experience some response, with 15–37% of
patients showing complete remission. These values are
significantly better than the results of nonlabeled rituxan
MoAb therapy alone. The duration of response ranges
from 0.5 to 24.9 months.

I-131 Tositumomab (I-131 Bexxar)

I-131 Tositumomab (I-131 Bexxar) is a murine IgG2a
Table 9-14 Radiation Dosimetry of Y-90 monoclonal antibody developed to target CD20 which is
Ibritumomab Tiuxetan (Zevalin) the same target for Y-90 zevalin. There are several similar-
and I-131 Tositumomab (Bexxar) ities and differences between Y-90 zevalin and I-131
Bexxar.
Organ Y-90 Zevalin T-131 Bexxar Indication
(mGy/MBq) In-111 Zevalin (mGy/MBq) I-131 Bexxar is not a first line therapy. It is recom-
Spleen 9.4 0.9 1.14
mended for CD20+ follicular NHL (with or without trans-
Testes 9.1 0.6 0.83 formation), which is refractory to rituximab.
Liver 4.8 0.7 0.82 Method
Colon 4.8 0.4 1.34 A protocol outline is provided in Table 9-16. Unlike
Heart 2.8 0.4 1.25 Y-90 zevalin, the I-131 Bexxar therapy dose can be
Thyroid 0.3 0.1 2.71
Lungs 2.0 0.2 0.79
imaged using the 364-keV gamma emissions. Whereas
Bladder 0.9 0.2 0.64 a therapy patient receiving zevalin will first receive In-
Kidney 0.1 0.2 1.96 111 zevalin for biodistribution imaging then Y-90 zevalin
Total body 0.5 0.1 0.24 for therapy, I-131 Bexxar is used for both the image and

Table 9-15 Therapy Protocol for Y-90 Ibritumomab (Zevalin)

I. Day 1 II. Biodistribution scans III. Therapy (Day 7-9)

Pretreatment: Rituximab First scan 2–24 hours Pretreatment Rituximab

250 mg/m2 250 mg/m2
< 4 hrs: Second scan 48–72 hours < 4 hours:
5 mCi (185 MBq) Y-90 Zevalin
In-111 Zevalin
Over 10 min 0.4 mCi/kg platelet >150k
0.3 mCi/kg platelet 100–150k
Optional:Third scan 90–120 hours
296 NUCLEAR MEDICINE: THE REQUISITES
Table 9-16 Therapy Protocol for I-131 Tositumomab (Bexxar)
Day 0 Day 1 Day 2, 3, or 4
Thyroid block 1. Premedicate: acetaminophen, Whole-body
diphenhydramine dosimetry scan
SSKI: 4 drops 2. Pretreatment Tositumomab
TID OR (450 mg) over 60 min
potassium
iodide:
130 mg QD
Note: begin day 3. I-131 Bexxar (5 mCi or 185 MBq)
before and over 20 min
continue for
2 weeks
4. Dosimetric whole-body
scan prevoid
therapy. The Bexxar regimen also uses the same tositu-
momab monoclonal antibody for pretreatment (in the
nonradiolabeled form) as for dosimetry and therapy (in
the I-131 labeled form). The I-131 label means that,
unlike a pure beta-emitter, the dose requires shielding.
Also, it must be determined before discharging the
patient that the exposure to others will not be greater
than 500 mrem from the patient.
One potential problem with I-131 Bexxar is the need
to perform additional scans after dosing to determine
dosimetry. This is less convenient than Y-90 zevalin,
where two scans rather than three or four are done to
examine biodistribution and not for dose calculations.
Because the I-131 radiolabel can disassociate from the
MoAb and result in unwanted thyroid exposure, the
patient must receive thyroid blocking medication at least
one day before beginning the studies. This usually
involves supersaturated potassium iodide (SSKI) four
drops three times a day or potassium iodide 130 mg
tablets once a day. This should be continued for 2
weeks.
The patient is first treated with nonlabeled tositu-
momab to block excess CD20 sites. This helps decrease
nonspecific antibody targeting. Following this, the
patient receives a low dosimetry dose of I-131 tositu-
momab. The first scan is done after infusion but before
the patient voids to determine the maximal initial activity
in the patient. Dosimetry scans are then done serially to
calculate how fast activity clears from the body (or resi-
dence time) before dosing can be calculated. Then a
patient-specific dose is calculated based on this informa-
tion as well as the desired dose based on platelet levels.
Toxicity
Like Y-90 zevalin, similar significant hematologic side
effects can occur. Up to 15% of patients may require sup-
portive care such as transfusions and colony-stimulating
Day 6 or 7 Day 7 (up to day 14)
Whole-body dosimetry scan Therapy administration
Calculate patient specific 1. Pretreat Tositumomab
dose: (450 mg) over 60 min
Plate >150k 75 cGy
2.Therapy dose I-131
Tositumomab over
20 min
Plate 100-150k 65 cGy
factor. Long-term side effects are possible, such as the
myelodysplastic syndrome and secondary leukemia.
Hypothyroidism may occur if proper premedication is
not given to block the thyroid from taking up I-131.
The other concern with any murine MoAb is the
HAMA response. HAMA is common initially with an inci-
dence that ranges widely in clinical use. The patients
who had extensive previous chemotherapy became
seropositive roughly 10% of the time, whereas those
patients who received I-131 Bexxar as a first-line therapy
had initially elevated titers up to 70% of the time.
Results
Overall, 63% of patients refractory to rituximab showed
response. Of these, 29% of the patients experienced
complete response. The median duration of response
was 26 months. This is significantly longer than with
Y-90 zevalin.
PALLIATION OF BONE PAIN
Metastatic disease to the bone is a common problem
causing significant pain and disability to patients with
cancer. Although this can occur with almost any cancer,
it most commonly involves tumors originating from
prostate, breast, and lung. Red marrow involvement is
most common, although cortical lesions are also fre-
quently seen. Numerous methods are available for the
treatment of bone pain. These include analgesics,
chemotherapy drugs, hormonal therapy, bisphospho-
nates, external beam radiation, and even surgery.
Radiopharmaceuticals are an important addition to this
list of treatments. Radiopharmaceuticals available for
treatment of bone pain are listed in Table 9-17.
Bone-seeking radiopharmaceuticals have been used to
treat bone pain from cancer for decades. These agents
 Oncology 297

Table 9-17 Radiopharmaceuticals for Bone Pain Palliation

Physical β Max β Mean Maximal tissue γ Photon

Radionuclide Pharmaceutical half-life (days) (MeV) (MeV) distance (mm) (keV)

P-32 Orthophosphate 14.3 1.71 0.695 8 –

Sr-89 Chloride 50.5 1.46 0.583 6.7 –
Sm-153 EDTMP 1.95 0.8 0.224 3.4 103
Re-186 HEDP 3.8 1.07 0.349 4.7 137

all localize to bone, in areas of bone repair and turnover.

Therefore, they deposit in areas of metastasis. The thera- Table 9-18 Dosimetry of Bone Pain Therapeutic
peutic effects depend on the emission of beta particles. Agents Strontium-89 (Sr-89) and
Beta particles are high energy but only travel millimeters Samarium-153 (Sm-153)
from the site of deposition. This ensures the effects are
limited to the abnormal bone and normal tissue is Sr-89 Sm-153
spared. These agents are extremely useful as they can be rad/mCi rad/mCi
Organ (cGy/37 MBq) (cGy/37 MBq)
given in addition to other therapies such as external
beam radiation, or even after external beam therapy has Bone surface 63.0 25.0
reached maximal limits. Red marrow 40.7 5.7
Colon 4.7 0.037
Bladder 4.8 3.6
Phosphorus-32 (P-32) Testes 2.9 0.02
Ovaries 2.9 0.032
P-32 is one of the earliest known bone-seeking radioiso- Whole body 2.9 0.04
topes. It has been used via intraperitoneal infusion for the
treatment of tumors such as ovarian cancer and in the treat-
ment of polycythemia vera. It is available for intravenous
administration for bone pain palliation. A range of skeletal
absorbed doses have been calculated (25–63 rad/mCi or Toxicity
0.68 –1.733 cGy/MBq). However,it appears that the normal Because toxicity occurs to marrow components
marrow receives a high dose relative to the tumor due to the platelets and white blood cells, the hematologic status of
distribution of P-32 in the inorganic matrix as well as the cel- each patient must be evaluated before therapy. After
lular regions. Also, the lack of a gamma emission means no obtaining a baseline platelet count, platelets should be
external imaging can be done to assess distribution. measured at least every other week. Typically, platelets
will decrease 30% from baseline and reach the nadir
between 12–16 weeks after therapy. Toxicity is gener-
Strontium-89 (Sr-89) ally mild; however, Sr-89 must be used with caution in
Sr-89 (Metastron,Amersham) is also a pure beta-emitter those with white cell counts less than 2400 and platelets
which has been approved by the FDA for the manage- less than 60,000. A small number of patients experi-
ment of metastatic bone pain under the trade name ence transient worsening of symptoms.
Metastron (Amersham). It selectively localizes in areas Approximately 20% of patients will become pain free.
of boney turnover, predominantly in blastic lesions. It is Roughly 75% of patients will experience some significant
retained longer in regions of metastasis than normal decrease in pain, although some series have reported up
bone. The pathway of excretion is predominantly to 90% of patients experience some symptom relief.
through the urine with about one third bowel excretion. Pain relief begins about 7–20 days after injection and gen-
Dosimetry is outlined in Table 9-18. erally lasts 3–6 months.
A 4-mCi (148 MBq) dose is administered intravenously
slowly over 1–2 minutes. An alternative dose of 55
uCi/kg (2.04 MBq/kg) may be used. Repeat dosing is Samarium-153 (Sm-153)
possible, but factors such as initial response, hematologic Sm-153 (Quadramet, DuPont) is a beta-emitting radio-
status, and current status must be considered in each pharmaceutical that has the added advantage of a gamma
case. In general, a repeat administration is not recom- emission that can be detected for external imaging. It
mended before 90 days have elapsed. has been approved for use in patients with osteoblastic
298 NUCLEAR MEDICINE: THE REQUISITES

Figure 9-31 Samarium-153 (Sm-153) palliation of metastatic disease bone pain. A, The whole
body Tc-99m MDP bone scan prior to therapy confirms the presence of osseous metastasis that will
accumulate the therapy agent. B, The whole body scan done with the Sm-153 therapy dose shows
close correlation with the lesions seen on bone scan.

metastases that can be visualized on a nuclear medicine
bone scan (Fig. 9-31).
It is administered in a 1.0-mCi/kg (37 MBq/kg) dose
intravenously over the course of 1 minute. Approximately
50% of the dose is localized to normal and abnormal bone.
It accumulates in metastatic lesion in a 5:1 ratio compared
with normal bone. Patients should be well hydrated and
void frequently as the primary route of clearance is
through the urine. Approximately 35% of the dose is
excreted in the first 6 hours.
As in Sr-89, the bone marrow toxicity is a limiting
factor. Toxicity is usually mild, although serious side
effects and even fatalities have been reported. Platelets
decreased on the order of 25% from baseline and white
blood cells by 20%.
The short range of the Sm-153 beta particle should be
advantageous when considering the dose to normal mar-
row. A response rate on the order of 83% has been
reported. Pain relief is generally noted within 2 weeks,
with a duration of 4–40 weeks.
Re-186 HEDP
Re-186 HEDP or rhenium-186 hydroxylidene diphospho-
nate (Rh-186 HEDP) is formed by combining a diphos-
phonate useful for bone pain therapy, etidronate, with
a beta emitter. Re-186 HEDP is another agent that may be
useful for the palliation of bone pain. It emits a gamma
ray useful for imaging and lesion identification. It rapidly
localizes to bone with approximately 14% retained in
 Oncology 299

bone. The remainder is rapidly cleared with roughly 70%
of the dose excreted in the urine 6 hours after injection.
LYMPHOSCINTIGRAPHY
Melanoma
Malignant melanoma can be cured by surgery if the dis-
ease is localized, but advanced disease is invariably fatal.
The main prognostic factors are thickness of the primary
lesion and metastatic lymph node involvement. Patients
with regional lymph node metastases have a 10% 10-year
survival rate; those with distant metastases have a less
than 1% 10-year survival.
Patients with a primary cutaneous melanoma thick-
ness of less than 1 mm have a greater than 90% 10-year
survival. Those with a primary lesion thickness greater
than 4 mm have a poor prognosis with a 4-year survival
of 40%, and the benefit of lymph node dissection is ques-
tionable. It is the patient with an intermediate thickness
lesion (greater than 1 mm but less than 4 mm) that can
benefit from lymph node dissection.
Lymph node metastases are found in approximately
20% of patients undergoing draining nodal bed resec-
tion. Complications of this surgery include chronic
lymphedema and nerve injury. Histopathologic evalua-
tion of the first lymph node draining the tumor, the sen-
tinel lymph node, predicts the presence or absence of
lymphatic spread and nodal involvement.
Lymphoscintigraphy can localize the sentinel lymph
node (Fig. 9-32) and determine the need for nodal basin
resection. Absence of tumor in the sentinel lymph node
has a very high negative predictive value because skip
metastases are rare (<2%). Sentinel lymph node biopsy is
associated with a very low incidence of late lymphedema
(<2%).
The pattern of lymphatic drainage is usually pre-
dictable when the lesion is located in the extremities.
Activity generally drains into axillary and groin lymph
node basins, although popliteal and epitrochlear nodes
are more common than previously thought. The pattern
of lymphatic drainage from lesions of the head, neck, and
trunk is much less predictable, sometimes crossing mid-
line, having multiple drainage patterns, and unusual path-
ways (e.g., the triangular intermuscular back space and
paravertebral and retroperitoneal nodal beds).
Vital blue dye has been used to detect the sentinel
node intraoperatively but this has limitations because of
its rapid transit. If surgery is not prompt, the dye may
move beyond the sentinel node to other nodes. Although
some surgeons still use this technique, many combine
both dye and radionuclide techniques. An intraoperative
gamma probe detector is often used to provide accurate
localization of the node at the time of surgery.

Figure 9-32 Melanoma lymphoscintigraphy. Filtered Tc-99m sulfur colloid injected

intracutaneously around the melanoma lesion. Note the early lymphatic drainage (small
arrowheads, upper left) at 10 minutes, first appearance of sentinel node (medium arrowhead,
upper right) at 20 minutes, and visualization of two nodes at 40 minutes (lower left) and 1 hour
(large arrowheads, lower right).
300 NUCLEAR MEDICINE: THE REQUISITES
Breast Cancer
The surgical management of breast cancer has evolved
over recent years from radical surgical procedures
toward lesser surgical procedures. With the success of
sentinel node identification for melanoma, surgeons
became interested in this approach to determine
whether a patient with newly diagnosed breast cancer
has regional tumor spread to axillary nodes.
Involvement of the regional nodal basin is the single
most important independent variable in predicting
prognosis. However, axillary lymph node dissection is
associated with considerable morbidity including
wound infection, seroma, paresthesia, and chronic limb
edema.
Increasingly, the sentinel lymph node approach has
gained favor with surgeons. Sentinel node lym-
phoscintigraphy using an intraoperative gamma probe
has greater than 90% accuracy for detecting sentinel
lymph nodes. A histopathologically negative node has
a very high negative predictive value for metastatic axil-
lary involvement.
Radiopharmaceuticals
Over the years, a number of radiopharmaceuticals have
been used for lymphoscintigraphy, including Tc-99m sul-
fur colloid (Tc-99m SC),Tc-99m human serum albumin
(HSA),Tc-99m nanocolloid, and Tc-99m antimony sulfur
colloid. Only Tc-99m SC is available in the United States.
Colloidal clearance rate depends greatly on particle size.
A large portion of unmodified larger Tc-99m sulfur colloid
particles are retained at the injection site. Therefore, the
Tc-99m sulfur colloid is first filtered using a 0.22-μm filter
to ensure a more uniform, smaller colloidal particle that is
more conducive to lymphatic drainage.
Methodology
Melanoma
The methodology of scanning in cases of melanoma
varies. Some surgeons order lymphoscintigraphy before
surgery and mark the location of the sentinel node on the
patient’s skin. Others do not request imaging but use
a radiation detector probe at surgery to find the sentinel
node, on which they then perform biopsy. The combina-
tion of the two methods is common.
Four to six injections of 100 μCi in tuberculin syringes
are made intracutaneously around the lesion or surgical
site. Sequential imaging is performed every 5 minutes
for 60–90 minutes until the sentinel node is detected.
The location of the node can then be marked on the skin
with an indelible marker pen.
Breast Cancer
Various injection methods have been investigated and
used on a clinical basis. One method uses an injection
around the tumor or biopsy site itself. This may require
ultrasound guidance for deep tumors. The lymphatic
channels are more limited in the deep subcutaneous tis-
sues compared with the dermal regions of the breast. In
addition, the lymphatics may be disrupted by the tumor
or surgery. Therefore, a dose injected in the peritumoral
location may not migrate. Some physicians use an intra-
dermal or subdermal injection overlying the tumor.
Another approach utilizes a periareolar injection that
takes advantage of rich lymphatic drainage. Although
this location may be distant from the tumor, lymphatic
drainage follows this pathway. The migration failure rate
using this periareolar method is extremely low. Finally,
a subareolar injection just proximal to the tumor or
biopsy site can be used. Care must be taken that activity
in the injection site will not obscure adjacent axillary
lymph nodes. The use of two methods together pro-
duces the best results.
On the day of surgery or even the day before surgery,
several injections of the radiopharmaceutical (around 100
μCi each) are made in a peritumoral location, proximal to
the biopsy site, or periareolar region. Imaging may or
may not be performed, although it is recommended to
visualize unusual routes of migration. Vigorous injection
site massage for 5 minutes improves the chance migration
will occur. An intraoperative gamma probe is used to
detect the sentinel node so that biopsy can be performed.
SUGGESTED READING
Alazraki NP, Eshima D, Eshima LA, et al: Lymphoscintigraphy, the
sentinel node concept, and the intraoperative gamma probe in
melanoma, breast cancer, and other potential cancers. Semin
Nucl Med 27:55-67, 1997.
Isreal O, Mor M, Epelbaum R, et al: Clinical pretreatment risk fac-
tors and Ga-67 scintigraphy early during treatment for predic-
tion of outcome of patients with aggressive non-Hodgkin’s
lymphoma. Cancer 94:873-878, 2002.
Krag D,Weaver D,Ashikaga T, et al: The sentinel node in breast
cancer: a multicenter validation study. N Engl J Med 339:941-
946, 1998.
Krenning EP, Kwekkeboom DJ, Bakker WH, et al: Somatostatin
receptor scintigraphy with [111 In-DTPA-Dphe-1] and [123I-
Tyr-3]-octreotide: The Rotterdam experience with more than
1000 patients. Eur J Nucl Med 20:716-731, 1993.
Kurtaran A, Scheuba C, Kaserer K, et al: Comparison of In-111
DTPA-D-Phe1-octreotide and Tc-99m-(V)-dimercapto-succinic
acid scanning in the preoperative localization of medullary thy-
roid carcinoma. J Nucl Med 39:1907-1909, 1998.
Mariani G, Gipponi M, Moresco L, et al: Radioguided sentinel
lymph node biopsy in malignant cutaneous melanoma. J Nucl
Med 43:811-827, 2002.

Moffat FL, Pinsky CM, Hammershaimb L, et al: Clinical utility of
external immunoscintigraphy with IMMU-4 technetium-99m
Fab′ antibody (Tc-99m CEA-SCAN) fragment in patients under-
going surgery for carcinoma of the colon and rectum: results of
a pivotal, phase III trial, J Clin Oncol 14:2295-2305, 1996.
Newman LA: Lymphoscintigraphy and sentinel lymph node
biopsy in breast cancer: a comprehensive review of variations in
performance and technique. J Am Coll Surg 199:804-816,2004.
Raj GV, Partin AW, Polascik TJ: Clinical utility of Indium-111-
Capromab pentreotide immunoscintigraphy in the detection of
early, recurrent prostate carcinoma after radical prostatectomy.
Cancer 94: 987-996, 2002.
Serafini AN: Current status of bone palliation with systemic
radioisotopes. Nucl Med Ann: 253-274, 2002.
Oncology 301
Taillefer R: The role of 99mTc-sestamibi and other conventional
radiopharmaceuticals in breast cancer diagnosis. Semin Nucl
Med 29:16-40, 1999.
Weiner RE,Thakur ML: Radiolabeled peptides in diagnosis and
therapy. Semin Nucl Med 31:296-311, 2001.
Willkomm P, Bender H, Bangard M, et al: FDG PET
and immunoscintigraphy with 99mTc-labeled antibody frag-
ments for the detection of colorectal carcinoma. J Nucl Med
41:1657-1663, 2000.
Witzig TE, Gordon LI, Cabanillas F, et al: Randomized controlled
trial of yttrium-90-labeled ibritumomab tiuxetan radioim-
munotherapy versus Rituximab immunotherapy for patients with
relapsed or refractory low-grade, follicular, or transformed B-cell
non-Hodgkin’s lymphoma. J Clin Oncol 20:2453-2463,2002.
 10
CHAPTER Oncology: Positron
Emission Tomography

Introduction and Background Testicular Carcinoma

Radiopharmaceuticals Prostate Carcinoma
F-18 Fluorodeoxyglucose Renal and Bladder Carcinomas
Investigational PET Imaging Agents Musculoskeletal Tumors
F-18 FDG PET Dosimetry
Protocol
Image Interpretation INTRODUCTION AND BACKGROUND
Normal Distribution and Variants of F-18 FDG
Benign Variants For decades, positron emission tomography ( PET ) has
Effects of Inflammation and Therapy been used for many research applications. In recent
Artifacts years, changes in reimbursement and availability have led
Patterns of Malignancy to the rapid expansion of PET for clinical patient care.
Clinical Uses of PET in Oncology The majority of these PET scans are performed to evalu-
Lung Carcinoma ate cancer. Uses include cancer diagnosis,staging,restag-
Diagnosis of Solitary Pulmonary Nodule ing and monitoring response to therapy ( Table 10-1).
Staging NSCLC In most cases, cancer cells are more metabolically
Restaging NSCLC and Assessing Response to Therapy active and divide more rapidly than normal tissues. By
Small Cell Lung Carcinoma using radiopharmaceuticals that target physiological
Head and Neck Carcinoma parameters such as glucose metabolism, PET enables
Thyroid Carcinoma imaging and quantification of cellular function and
Esophageal Carcinoma tumor detection. This approach has several potential
Diagnosis advantages over anatomic modalities like computed
Staging tomography (CT).
Restaging and Monitoring Response to Therapy CT imaging relies on size and architectural changes to
Colorectal Carcinoma diagnose malignancy, which limits sensitivity and speci-
Other Tumors of the Gastrointestinal Tract ficity. For example, when lymph nodes are identified in
Lymphoma patients with cancer, enlarged nodes are assumed to
Imaging harbor malignancy, whereas normal-sized nodes are char-
Detection acterized as benign. Therefore, nodes seen on CT are
Staging, Monitoring Therapy, and Restaging often inaccurately characterized as benign because they
Melanoma are not pathologically enlarged.
Breast Carcinoma Often, it is not possible to determine if a residual mass
Diagnosis seen on CT after therapy contains tumor, and a change in
Primary Breast Cancer the size of a mass does not accurately predict tumor
Lymph Node Evaluation response to therapy. Another limitation of CT is that
Clinical Use of PET tumor evaluation after therapy is difficult, and scarring
Ovarian Carcinoma from surgery and radiation may obscure malignant dis-
Cervical Carcinoma ease. PET, on the other hand, permits monitoring of

302
 Oncology: Positron Emission Tomography 303

Table 10-1 Historical Perspective on Medicare-approved Indications of F-18 FDG PET in Oncology

Clinical condition Effective date Coverage

Solitary pulmonary nodule January 1998 Characterization

Lung cancer (non-small cell) January 1998 Initial staging
Lung cancer (non-small cell) July 2001 Diagnosis, staging, restaging
Esophageal cancer July 2001 Diagnosis, staging, restaging
Colorectal cancer July 1999 Locating tumors if rising CEA suggests recurrence
Colorectal cancer July 2001 Diagnosis, staging, restaging
Lymphoma July 1999 Staging and restaging (only when used as an
alternative to Gallium scan)
Lymphoma July 2001 Diagnosis, staging, restaging
Melanoma July 1999 Evaluate recurrence prior to surgery instead of
Gallium scan
Melanoma July 2001 Diagnosis, staging, restaging (evaluating regional
nodes not covered)
Breast cancer October 2002 Staging when distant metastasis, restaging
locoregional recurrence or metastasis, monitoring
treatment response for locally advanced and
metastatic cancer when change in therapy is
anticipated
Head and neck cancers (excluding July 2001 Diagnosis, staging, restaging
central nervous system and thyroid)
Thyroid cancer October 2003 Restaging recurrent or residual cancers of follicular
cell origin that were previously treated by
thyroidectomy and radioiodine ablation with
serum thyroglobulin >10 ng/ml and negative
I-131 whole body scan
Cervical cancer Under CMS consideration for Possible detection of residual or recurrent tumor
coverage after therapy
Ovarian cancer Under CMS consideration for coverage Initial staging, detecting recurrence, detecting
recurrence when CA-125 titre is rising but
computed tomography is negative
Pancreatic cancer Under CMS consideration for coverage Possibly differentiating benign from malignant,
detect metastases
Small cell lung cancer Under CMS consideration for coverage Initial staging, restaging, diagnosis of occult tumor
in paraneoplastic syndrome
Testicular cancer (pure seminoma or Under CMS consideration for coverage Initial staging, evaluate residual mass, recurrence
nonseminomatous germ cell) when computed tomography is normal but serum
factors rising

CEA, Carcinogen embryonic antigen; CMS, Centers for Medicare and Medicaid.

activity levels on serial studies. Changes in metabolic

activity better characterize a mass and better predict ther-
apy outcome than anatomic measurements of size.
However, PET is limited by a lack of anatomic detail.
Normal uptake in structures like bowel, muscles, and
ureters can be mistaken for tumor. Therefore, correla-
tion with CT or magnetic resonance imaging ( MRI ) is
critical for image interpretation. If the CT is performed
separately from the PET, comparison of the two data sets
can be done side-by-side or after processing with image
fusion software. Increasingly, PET is acquired sequen-
tially with the CT on a PET-CT scanner. This allows the
most precise image fusion as differences in position are
minimal. The combination of anatomic information with
metabolic data is the most accurate method for evaluat-
ing malignancy.
Radiopharmaceuticals
Many commonly used positron-emitting radioisotopes
are based on atoms found in organic substances: oxygen-
15, nitrogen-13, carbon-11, and the hydroxyl analog, fluo-
rine-18 ( Table 10-2). Short half-lives limit the clinical
usefulness of many positron emitters,requiring a cyclotron
to be in close proximity. However, F-18 has a 110-minute
half-life and can be delivered from regional, commercial
cyclotrons.
The various positron emitters can be attached to vari-
ous biological carrier molecules. Carrier molecules such
as nucleosides, amino acids, fatty acid components, and
glucose analogs are chosen to form radiopharmaceuti-
cals that target components of cellular metabolism and
304 NUCLEAR MEDICINE: THE REQUISITES

Table 10-2 Physical Characteristics of Important

Positron-emitting Isotopes in Oncology

Radio- Half-life Decay β+Emax Range

isotope (min) mode (%) γ (keV) (MeV) (mm)

O-15 2.07 β+ (99.9) 511 1.72 8.0

EC (0.1)
N-13 9.96 β+ (100) 511 1.19 5.4
C-11 20.4 β+ (99.8) 511 0.96 4.1
EC (0.2)
F-18 109.7 β+ (97) 511 0.635 2.4
EC (3)

β+, positron emission; EC, electron capture.

Figure 10-1 Glucose and F-18 FDG intracellular kinetics.

division. Targets include DNA synthesis, membrane syn-
thesis, and glucose metabolism ( Table 10-3). The chem-
istry of labeling carrier molecules with positron emitters
is usually much simpler than labeling with gamma emit-
ters such as technetium-99m. Currently, the only PET
radiopharmaceutical approved by the Food and Drug
Administration for clinical use in oncologic imaging is
F-18 fluorodeoxyglucose ( F-18 FDG).
F-18 Fluorodeoxyglucose
Cancer cells generally have a higher level of metabolic
activity than normal tissues and use more glucose. In
malignant cells, higher levels of glucose membrane trans-
porters increase intracellular glucose uptake. Within the
cancer cell, hexokinase ( hexokinase II) activity levels are
increased, and the phosphorylated glucose then moves
through the glycolysis pathway.
F-18 FDG is a glucose analog that is taken into the cell
and phosphorylated by the same mechanism as glucose.
Increased F-18 FDG activity is seen in tumors for several
reasons. First of all, increased glucose transporter activ-
ity is present. In addition, glucose-6-phosphatase levels
are low in cancer cells, and the phosphorylated FDG can-
not diffuse out of the cell. Unlike glucose,F-18 FDG is not
metabolized further and remains trapped (Fig. 10-1).
Table 10-3 Carrier Molecules for PET Imaging in
Oncology
Agent Target
Deoxyglucose Glucose metabolism
Thymidine DNA synthesis
Acetate Lipid synthesis
Choline Lipid synthesis, membrane
synthesis
Tyrosine Protein synthesis
Methionine Protein synthesis
After intravenous injection, F-18 FDG rapidly distrib-
utes throughout the body. Cellular uptake and
phosphorylation occur as background activity clears.
The primary route of radiotracer excretion occurs by the
kidneys, although F-18 FDG excretion also takes place
through the bowel. Optimal imaging time is between 40
to 60 minutes, based on maximum uptake, background
clearance, and physical half-life.
Many factors affect F-18 FDG uptake, distribution, and
clearance. Serum glucose actively competes with F-18
FDG uptake. Insulin occurring endogenously after eat-
ing or administration to diabetics will increase uptake to
the liver and soft tissues, thereby decreasing uptake in
tumors. Also, inflammation and infection may result in
uptake that rivals or exceeds that of a malignancy.
Investigational PET Imaging Agents
F-18 FDG uptake in tumor is not specific for malignancy.
Clinically, it is often critical to differentiate cancer from
inflammation. Many radiotracers that are more specific
for malignancy by imaging the increased cellular division
seen with cancer are under investigation. Other agents
may be superior to F-18 FDG because of better tumor
uptake or reduced background activity.
The radiolabeled nucleoside F-18 fluorothymidine (F-18
FLT ) is a marker of cellular proliferation. It has shown
a promising ability to predict tumor grade in lung cancer
evaluation,better evaluates many brain tumors,and may be
Label
an earlier predictor of tumor response. C-11 methionine,
F-18 an amino acid, has accurately evaluated various cancers
F-18, C-11 including brain tumors. C-11 choline and C-11 acetate
F-18, C-11 have shown promise in prostate cancer that may be due,in
C-11
part, to the lack of urine activity in bladder adjacent to the
F-18, C-11 prostate bed as well as differences in uptake activity. F-18
C-11 misonidazole is a marker of tumor hypoxia and may
play a role in therapeutic decision-making. Additional PET
 Oncology: Positron Emission Tomography 305

radiopharmaceuticals are being developed that bind to

receptors, including estrogen receptors in breast carci- Box 10-1 Protocol for F-18 FDG Imaging
noma. These agents are not available for clinical use. in Oncology
In addition to these newer agents, F-18 sodium fluoride
(F-18 fluoride) is gaining renewed interest. F-18 fluoride PATIENT PREPARATION
was originally used as a bone-scan agent, and it may be Avoid strenuous exercise for several days
superior to Tc-99m MDP or F-18 FDG in the evaluation of Diabetics well controlled, take medication on day of
several skeletal tumors. This includes highly lytic tumors scan but no insulin within 2 hours of FDG injection
such as multiple myeloma. NPO 4–6 hours or overnight, avoid carbohydrates
Check serum glucose (<180–200)
F-18 FDG PET Dosimetry Patient kept warm and relaxed
Radiation dosimetry values for F-18 FDG are listed in Consider sedation for: claustrophobia, anxiety/tense
Table 10-4. The whole body effective dose from a standard muscles, cancer in head and neck, prior brown fat
uptake
400-MBq scan is 1.6 mSv (160 mrem). The effective radia-
tion dose of a low-dose whole-body CT performed for PET- DOSE
CT varies, but may be 2 mSv (200 mrem). This compares Adults: 10–15 mCi (0.21 mCi/kg) intravenously
to a typical diagnostic chest CT,in which the effective dose Children: 150 uCi/kg
frequently reaches 8 mSv due to increased radiation used Wait 45-60 minutes, patient to avoid movement and
to visualize structures optimally. speech
Void

Protocol ACQUISITION AND PROCESSING

Field of view
An example of a protocol for F-18 FDG PET in oncology
Varies by patient size (50 cm)
patients is included in Box 10-1. Because glucose com-
petes with F-18 FDG, protocols include measures to limit Transmission scan
the impact of serum glucose of the scan. Patient prepara- Stand-alone PET
tion generally includes fasting overnight or for at least 4-6 External rod source
hours before injection and avoiding carbohydrates in the Minute/bed position
meal before injection. The serum glucose level of patients
PET-CT
should be checked before injection. The upper-limit cut-
off varies among institutions,but a value under 200 mg/dl is Low-dose CT (70–80 mA, 140 kvP)
generally considered acceptable. Diabetics are asked to Emission scan
follow the same preparation routine and take their insulin. 5–10 minutes/bed position
However, insulin should not be given within 2 hours of Repeat imaging each section or bed position until area
radiotracer injection. Non–insulin-dependent diabetics are covered
treated in a similar fashion.
Claustrophobic patients may require sedation, particu- Processing
larly when a PET-CT is performed because the machine has Filtered backprojection or iterative reconstruction
a deep bore. Sedatives and beta-adrenergic blockers are

Table 10-4 F-18 FDG Dosimetry sometimes used to decrease uptake in the supraclavicular
fat (so-called “brown fat”), although this is often with lim-
Organ rad/mCi (cGy/37 MBq) ited effectiveness. Patients should be kept warm, quiet,
and relaxed during the injection and uptake phases to
Bladder wall 0.32 decrease muscle and brown-fat uptake. Patients should be
Myocardium 0.22
instructed to avoid strenuous exercise for a couple of days
Brain 0.07
Liver 0.058 before the study.
Kidneys 0.074 The standard F-18 FDG dose is 10–15 mCi (370–555
Red marrow 0.047 MBq) intravenously. An absorption and clearance period
Testes 0.041 necessitates a delay of 40–60 minutes before scanning.
Ovaries 0.053
Immediately before being placed on the scanner, patients
Colon 0.046
must void.
Critical organ in bold. Patients are usually imaged in a supine position. Because
From package insert for MetaTrace, PETNET Pharmaceuticals, Knoxville,TN. attenuation correction with CT causes beam-hardening
306 NUCLEAR MEDICINE: THE REQUISITES
artifact when the arms are in the field of view,arm position
must always be taken into consideration. Arms are placed
above the head when the pathology is in the chest,
abdomen, and pelvis. In cases where the tumor is in the
head and neck,patients are imaged with arms at their sides.
Scanning usually begins at the head but will begin at the
thighs when tumor is in the pelvis to minimize the impact
of urine activity accumulating in the bladder.
The imaging is done in two stages. First, a transmis-
sion scan using an external positron or x-ray source is
performed for attenuation correction. The positron
emission scan is then acquired, which detects the pho-
tons from the F-18 FDG. As the administered dose is
known and can be time-decay corrected, the number of
photons striking the detector will reflect the metabolic
activity of a lesion once a correction for differences in tis-
sue attenuation is applied. Therefore, attenuation correc-
tion allows the levels of activity in the patient to be
accurately quantified. An attenuation correction factor is
calculated by comparing the counts striking the detector
from the transmission scan through the patient to a blank
scan where no patient is present: attenuation correction
factor = (counts blank scan) / (counts transmission scan).
External rods ( germanium-68 or cesium-137) rotating
around the patient can be used for this attenuation cor-
rection transmission scan. The process requires 4–6
minutes per image level (or bed position) with a typical
whole-body scan requiring 5–7 bed positions. A PET-CT
scanner uses the CT transmission data for attenuation
correction, as well as for image fusion for anatomic local-
ization. The CT takes only seconds to complete.
The emission scan data are then acquired from the F-18
FDG activity emitted from the patient. Depending on
patient size and radiopharmaceutical dose, the emission
scan takes 5–10 minutes per bed position. A whole-body
scan has traditionally been considered a scan from skull
base to mid-thigh. Such a whole-body scan will require
approximately 35–40 minutes. If the entire brain or the
lower extremities are included, more bed positions are
added.
Currently, most scans done on a PET-CT are performed
without intravenous or oral contrast. Contrast adminis-
tration requires additional time, support personnel,
equipment, and patient supervision. Also, CT contrast
agents cause increased attenuation on the transmission
scan, which can lead to areas of artifactually increased
activity on attenuation-corrected PET images. Using
dilute oral contrast or water minimizes the impact of this
artifact. Performing the attenuation-correction CT scan
after the arterial phase of the intravenous contrast bolus
will reduce the artifact from intravenous contrast. In
practice, examining the nonattenuation-contrast images
should allow an experienced reader to avoid confusion
caused by contrast artifact. Newer software may
decrease the artifacts from contrast.
Following the scan, reconstruction is performed.
Although ititerative reconstruction is becoming the norm,
some scanners will use filtered backprojection, including
for nonattenuation-corrected images. Data is displayed as
a maximal intensity projection (MIP) rotating image and/or
as transaxial, coronal and sagittal slices. Both attenuation-
corrected images and nonattenuation-correction images
should be reviewed as artifacts and lesions may be more
obvious in one or the other. PET-CT scanners will automat-
ically provide images fused to the CT in each orthogonal
plane. If the CT is done in a separate imaging session, pos-
timage fusion software may require manual alignment or
may offer semiautomatic capabilities. A current correlative
CT is needed for optimal interpretation.
IMAGE INTERPRETATION
Normal Distribution and Variants
of F-18 FDG
The normal distribution of F-18 FDG reflects glucose
metabolism ( Fig. 10-2). The brain is an obligate glucose
user,so uptake is high. The kidneys,ureters,and bladder also
show intense activity due to urinary clearance of F-18 FDG.
Moderate and sometimes heterogeneous activity is seen in
the liver. Variable activity occurs in the heart, gastrointesti-
nal tract, salivary glands, and testes. The uterus may show
endometrial uptake depending on the menstrual cycle stage.
Low-level activity is normal in the bone marrow.
The urinary activity can lead to interpretation difficul-
ties. Although the ureters usually appear as long tubular
structures, they may be seen as very focal areas of activity
that may be confused with pathology. CT correlation or
PET-CT can help localize the activity to a visible ureter or
show that no mass or lymph node is present. F-18 FDG
in the bladder and kidneys can prevent visualization of
tumors in those structures. The bladder may also limit
evaluation of other tumors in the pelvis.
When imaging tumors located near the heart, minimiz-
ing cardiac activity is desirable. The myocardium uses
glucose as an optional fuel. In a fasting state, fatty acid
metabolism dominates over glucose use, leading to
decreased FDG uptake. However, fasting yields inconsis-
tent results. In fact, significant cardiac uptake is seen in up
to 50% of fasting patients. Myocardial uptake is usually not
seen in the right ventricle and may be heterogeneous in the
left ventricle.
Normal excretion of F-18 FDG is highly variable
throughout the gastrointestinal tract. Low-level activity
can be seen focally or diffusely in the esophagus. In gen-
eral, this normal uptake is less than that seen with
esophagitis or cancer. Intense activity in the stomach
sometimes limits the use of F-18 FDG PET in the evaluation
of gastric adenocarcinoma and gastric lymphoma. Activity

Figure 10-2 Normal distribution of F-18 fluorodeoxyglucose:
F-18 FDG uptake is normally intense in the brain and urinary tract,
moderately intense in the liver, and variable in muscles (especially
of the oropharynx), heart, and bowel.
in both large and small bowel is especially problematic as it
may obscure tumor in the bowel and mesentery ( Fig.
10-3A). Fusion of the PET and CT often helps localize focal
uptake to bowel wall or mesentery. Focal accumulation of
F-18 FDG may be caused by malignant and villous adeno-
mas,but it is frequently a normal,transient finding.
Activity in the oropharyngeal cavity is highly variable.
Low-level activity is normally seen in the salivary glands.
Nonspecific diffuse uptake is occasionally seen in the
parotid glands bilaterally in patients undergoing therapy.
Focal, intense uptake is often seen in oropharyngeal lym-
phoid tissue, including palatine and lingual tonsils ( Fig.
10-3B). Although it is most often symmetric, asymmetry
can occur normally or due to therapy. This may make it
difficult to evaluate tumor.
Benign Variants
Many processes alter F-18 FDG distribution ( Box 10-2)
and may affect the scheduling of a PET scan ( Box 10-3).
Oncology: Positron Emission Tomography 307
In addition, there are many reasons for the radiotracer
distribution to be altered in clinical use.
Activity is frequently seen in muscles that can some-
times pose interpretative problems. The muscles of the
oropharynx will show variable activity, and patients
should avoid speaking after injection to decrease artifact.
The skeletal muscles may show prominent uptake due to
recent exertion and tension. Insulin administered to dia-
betics or increased endogenous insulin occurring after
eating may also cause intense levels of muscle activity
necessitating a repeat exam (Fig. 10-4). Unilateral uptake
may occur in a normal vocal cord in a patient with con-
tralateral vocal cord paralysis ( Fig. 10-5 and 10-A [see
color insert]) or when surgery, such as for head and neck
cancer, causes contralateral muscles to be unbalanced.
One interesting and common variant is supraclavicular
F-18 FDG accumulation in so-called “brown fat”(Figs. 10-6
and 10-B [see color insert]). This activity was originally
thought to occur in the muscles of the neck, but fused
PET-CT images showed the uptake actually localized to
areas of fat on the CT scan. This variant occurs more
commonly in cold weather and cold patients, which
relates to the origin of the tissue as a primitive, nonshiv-
ering warming mechanism. Brown fat contains adrener-
gic receptors that contribute to uptake in anxious
patients. Although the pattern is simple to recognize, it
may decrease sensitivity for tumor detection in the
region. Variable degrees of improvement have been
obtained with sedation and adrenergic blocking agents
in anxious patients or those with a previously abnormal
scan. Warming the patient before injection and keeping
them warm is most effective in diminishing this uptake.
Focal F-18 FDG accumulation may localize to vessels,
particularly in the aortic arch. This may be associated
with calcifications from atherosclerotic disease. In gen-
eral, this activity is nonspecific, although intense uptake
could indicate arteritis.
The thyroid gland may show different uptake patterns
( Fig. 10-7). Diffusely increased activity may be seen in
thyroiditis, goiter, and Graves’ disease. The significance
of low-level diffuse uptake in patients without identifi-
able thyroid disease is uncertain; it may be normal or the
result of subclinical thyroiditis. Focal uptake can be seen
in benign nodules. However, focal activity can be the
result of malignancy in up to 50% of cases and evaluation
is warranted.
Benign adenomas and tumors outside the thyroid
gland may also accumulate F-18 FDG. This includes ade-
nomas in colon, parotid gland, and benign ovarian
tumors.
Effects of Inflammation and Therapy
F-18 FDG uptake is not specific for tumor. Increased activ-
ity can be seen in inflammation and infection, and
the cause has been attributed to glycolytic activity in
308 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-3 Normal variants. A, Normal intense uptake can be seen in small or large bowel.
B, Axial PET and corresponding CT images show normal uptake in the oropharynx. Normal uptake
is often symmetric and may be very intense when patients swallow excessively or talk. L, Lingual
tonsils; M, mandible; Mx, maxilla; P, parotid gland; PT, palatine tonsils; S, submandibular gland.

Box 10-2 Factors Affecting F-18 FDG Box 10-3 Clinical Factors Altering Patient
Uptake Scheduling

INCREASE UPTAKE History Course of Action

Higher tumor grade
Prior surgery Delay scan 4–6 weeks
Large number of viable cells
Chemotherapy Delay scan several weeks or
Increased tumor blood flow
schedule scan just before
Inflammation
next cycle
Tumor hypoxia
Radiation therapy Delay scan at least 3 months
Acute radiation
Colony stimulating Delay scan 1 week for short
Acute chemotherapy
factor (G-CSF) acting agents or several
Recent surgery
weeks for long acting
Serum glucose >200 Reschedule scan to control
DECREASE UPTAKE glucose
Benign lesion Insulin administration Wait at least 2 hours
Necrosis Breastfeeding Discontinue feeding at least
Low-grade or low-cellularity tumor, mucinous tumors, 6 hours
bronchoalveolar carcinoma
Hyperglycemia
High insulin
Chronic radiation leukocytes. Infections such as pneumonia will have
Prior chemotherapy intense radiotracer accumulation. Inflammatory uptake in
Scar a lymph node or mass cannot be differentiated from malig-
nancy. Such uptake may be problematic in sarcoidosis and
granulomatous disease in the chest ( e.g., histoplasmosis
 Oncology: Positron Emission Tomography 309

and tuberculosis). Other inflammatory processes in the

Figure 10-4 Muscle activity from increased insulin levels.
Intense, diffuse uptake in the muscles caused by eating prior to
radiotracer injection might obscure malignancy and require a
repeat exam.
lungs, such as occupational lung diseases and active inter-
stitial fibrosis and pneumonitis, may also cause markedly
abnormal uptake (Fig. 10-8).
Increased activity around a joint in the soft tissues or joint
capsule may be confused with a metastatic lesion. This pat-
tern is most common in the hip and shoulder. Fused PET-CT
images can help localize the uptake. Activity involving the
joint surface or both side of the joint may be present with
degenerative disease. Acute and healing fractures normally
accumulate F-18 FDG (Fig. 10-9). Correlation with the CT
can usually help differentiate this from a metastatic lesion,as
the fracture will be seen radiographically.
Therapy often results in an inflammatory response caus-
ing increased activity ( Figs. 10-10 and 10-11). There are
no hard and fast rules on how long to wait following ther-
apy to perform a PET scan. At times, repeat or even serial
imaging may be needed to confirm that increased activity is
iatrogenic. Radiation therapy causes intense F-18 FDG
uptake acutely. Because this uptake may persist for many
months, delaying the PET scan for 3 months is recom-
mended. Chemotherapy may cause a lesion to show
a transient apparent worsening. A delay in scanning of sev-
eral weeks or until just before beginning the next
chemotherapy cycle is currently recommended. However,
increasing data indicates that imaging early can predict
therapeutic response. The postoperative inflammatory

Figure 10-5 Vocal cord paralysis.A, Maximum-intensity projection whole-body image for restaging
lung carcinoma reveals suspicious focal right neck.Faint posttherapy change is seen in the right upper
lobe from radiation.B, Axial CT and PET images localize the neck uptake to the right vocal cord,ruling
out malignant adenopathy. Although therapy was to the right chest,mediastinal radiation affected the
recurrent laryngeal nerve on the left. The paralyzed left vocal cord has no uptake.
310 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-6 Brown fat uptake. A, CT and axial PET images show intense supraclavicular activity
in the fat. B, Supraclavicular activity can also be caused by malignant adenopathy as in this patient
with lymphoma or benign muscular activity. Fusion images can help clarify the etiology.
C, Commonly, whole-body PET may show benign intense costovertebral uptake in these same
patients, as well as the supraclavicular brown-fat activity.

Figure 10-7 Patterns of thyroid activity. A, Diffuse thyroid uptake suggests a benign process as in this
patient with Hashimoto’s thyroiditis, although the etiology of uptake is not always known. B, Focal
uptake due to a benign right adenoma in this patient will have a similar appearance to a malignant nodule
so follow-up is needed.
response in the wound healing process results in F-18 FDG
uptake that is usually mild to moderate ( Fig. 10-12) and
a delay of at least 2–4 weeks to minimize the effects of
inflammation on uptake is recommended.
Evaluation of the bone marrow may be limited by the
effects of therapy. When marrow-stimulating drugs are
used on patients with anemia or undergoing chemother-
apy, a diffuse increase in radiotracer uptake may result.
Usually this pattern is easily differentiated from metastatic
disease ( Fig. 10-13). However, increased marrow back-
ground can mask actual lesions from tumor involvement.
If at all possible,scans should be delayed until the effects of
marrow-stimulating therapy has resolved. In short-acting
agents,this typically takes 5–7 days but it is prolonged with
newer,long-acting drugs.
Low-level activity may be seen in the thymus in young
patients. Activity in the anterior chest may have a character-
istic shield shape of the thymus (Fig. 10-14). In cases
where activity appears following therapy,careful correlation
with the CT can help determine if normal-appearing thymic
tissue is present in the anterior mediastinum. Increased
uptake following therapy is known as “thymic rebound”and
can be seen on gallium scanning as well. It may be difficult
to differentiate from tumor involvement and response.
Artifacts
When metal or dense-iodinated contrast is present,
the attenuation-correction images may mistakenly show
increased radiotracer activity around the area (Fig. 10-15).
Examining the nonattenuation-corrected images, which
Oncology: Positron Emission Tomography 311
will show no increased activity, can lead to the correct
interpretation. Correlation with a current CT should be
done to help identify sources of such artifact.
Areas of intense F-18 FDG activity, such as in the blad-
der and infiltration at the injection site, can cause
a reconstruction artifact manifested by a band of artifac-
tually decreased activity across the patient. This was
more common with older systems that relied on filtered
backprojection and is less of a problem with iterative
reconstruction. Again, the nonattenuation-corrected
images show less effect.
Although combined PET-CT scanners have been a major
advance, these hybrid systems can generate certain arti-
facts. A common artifact is caused by misregistration of
PET and CT data due to respiration. A CT acquired with
breath-holding will greatly alter the position of organs com-
pared with the PET, which must be done in quiet respira-
tion. If the CT is done in quiet respiration,structures more
closely match the PET, but motion and low lung volumes
may obscure lesions. Respiratory motion artifact can also
cause abnormal uptake from a lesion to appear in an incor-
rect location on the CT, particularly for pathology near the
diaphragm. This most frequently involves a liver lesion
projecting over the lung or rib on the CT (Fig. 10-16). If
the patient is large or imaged with arms at their sides, trun-
cation artifact may lead to thin linear bands of activity run-
ning the length of the patient on the maximum-intensity
projection image. CT beam-hardening artifact is a com-
mon problem that affects the quality of the CT and fused
PET-CT images ( Fig. 10-17). This can be minimized by
312 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-8 Abnormal activity with inflammatory processes on PET. A, CT ( left) reveals severe
changes from occupational lung disease. Uptake is seen in areas of active inflammation on axial
(middle) and coronal (right) PET images. FDG PET does not differentiate tumor from benign
inflammatory change. B, Bilateral hilar uptake is seen on the coronal PET image of a patient with
marked adenopathy worrisome for lymphoma. Biopsy later proved this to be secondary to
sarcoidosis.
 Oncology: Positron Emission Tomography 313

Figure 10-9 F-18 FDG uptake in fracture. A, CT shows a left rib fracture (arrow) following biopsy of
a lung cancer. B, PET shows uptake in the fracture as well as the left suprahilar mass, which is not well
seen on the single noncontrast CT slice.

Figure 10-10 FDG posttherapy uptake. CT and PET images show acute radiation changes in the
posterior medial left lung. The uptake may decrease on follow-up scans.

moving the arms out of the field of view. Another common,
subtle artifact is a thin horizontal band or seam perpendicu-
lar to the patient’s axis from adjoining bed positions.
Patterns of Malignancy
Usually, there is a greater degree of uptake in more
aggressive tumors due to higher levels of metabolic activ-
ity. This pattern must be differentiated from the intense
activity often seen in infection or following radiation
therapy. Low-level activity may be seen in low-grade
tumors and tumors with lower relative numbers of cells
such as bronchoalveolar carcinoma and mucinous adeno-
carcinoma. Malignant pleural effusions most often have
low-level F-18 FDG activity and some are even negative,
which may be due to the dispersion of tumor cells in the
fluid so uptake is not detected.
Areas of tumor necrosis will have diminished F-18 FDG
accumulation. This is often seen as absent activity cen-
trally in very large masses. By determining areas of necro-
sis and intense activity, PET scans can help direct biopsy
for more sensitive accurate sampling. It may not be possi-
ble to differentiate a cavitary infectious process from
a necrotic tumor on PET, as both will have a cold center
and a peripheral rim of increased activity (Fig. 10-18).
Levels of background activity play a role in the detec-
tion of malignant lesions. For example, the high back-
ground activity of the brain limits sensitivity for metastatic
disease, with perhaps only a third of lesions being visual-
ized. Also, if background uptake is heterogeneous, as may
happen in the liver,it can make lesion detection more diffi-
cult. It is helpful to describe or grade the severity of
abnormal activity in terms of lesion to background differen-
tial. For example,lymph node activity in the hila and medi-
astinum are compared to the background mediastinal
activity. The uptake can be graded as mild, moderate, or
severe depending on the level above normal adjacent tissue.
Lesion activity can be quantified with a lesion-to-back-
ground ratio, lesion-to-liver ratio, or with a standard
uptake value ( SUV ). Quantification can help confirm
the visual impression and help follow abnormalities.
The SUV is a measure of the relative uptake in a region of
interest. It is corrected for body mass or body surface
area. F-18 FDG distribution is very low in fat that leads to
higher values in tumor and normal tissues in heavier
patients than thin patients.
314 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-11 Positive PET scan caused by pleurodesis.Maximal-intensity projection (A) images and
PET and CT images (B) show thickened left pleura on the CT (left) and intense FDG uptake on attenuation
corrected images (middle). This uptake may be difficult to differentiate from tumor. Note the typical
differences on the nonattenuation-corrected image (right) where the lungs appear “hotter,” as does the skin.

tissue activity (mCi/ml)

SUV =
injected dose (mCi) / body surface area
In general, an SUV >2.5 is considered suspicious for
malignancy. Most tumors have an even higher SUV.
There is a great deal of overlap with inflammatory
processes. Numerous factors affect SUV levels (Box 10-4).
For example, when comparing serial exams, different
imaging at different times after injection may alter SUV
values. Lesion size is also an important consideration.
Volume averaging can artifactually lower SUV values, as
regions of interest may include pixels from normal sur-
rounding tissue in small tumors or from motion.
CLINICAL USES OF PET IN ONCOLOGY
The use of PET scanning in primary tumors of the brain is
discussed in Chapter 13. Evaluation of lung cancer in
solitary pulmonary nodules was the first approved clini-
cal indication for F-18 FDG PET scanning in the U.S by
the Centers for Medicare and Medicaid ( CMS). Since
that time, the number of applications for PET has
increased. Although the list continually changes, indica-
tions for F-18 FDG PET approved by the CMS are listed in
Table 10-1. New indications are continually being evalu-
ated and added to the list, and private insurers may cover
other indications. Box 10-5 lists tumors that show a low
degree of F-18 FDG uptake, causing a lower sensitivity.
Lung Carcinoma
Lung carcinoma is the most common malignancy and has
the highest cancer-related death rate. The histologic
classification of lung cancer is outlined in Box 10-6.
Non–small-cell lung carcinoma ( NSCLC) accounts for
roughly 80% of cases and small-cell lung carcinoma
 Oncology: Positron Emission Tomography 315

Figure 10-12 Postoperative change. A, Two months after laparotomy, the CT shows secondary
changes in the midline anterior abdominal wall and stranding of the left lower quadrant peritoneal
fat.The corresponding PET image has mild anterior soft tissue uptake as well as normal uptake in
bowel and marrow. B, More intense uptake can be seen in the sagittal image from 6 months earlier.

(SCLC) accounts for the remaining 20%. Approximately
75% of SCLC cases present with disseminated disease.
Therefore, surgery is rarely indicated and chemotherapy
and radiation therapy are used. NSCLC, on the other
hand, is often resectable. Early diagnosis and proper
staging are critical to therapeutic planning in NSCLC.
Presenting clinical and radiographic findings are vari-
able in lung cancer. Patients may be asymptomatic or
experience hemoptysis, cough, weight loss, and symp-
toms of metastatic disease. Radiographic findings are
also nonspecific. A mass with an irregular, spiculated
border is malignant in up to 85% of cases, but lesions
with smooth margins may be cancerous over a third of
the time. Workup for patients with these abnormal
radiographs might include sputum cytology, bron-
choscopy, transthoracic needle biopsy, and medi-
astinoscopy. Each of these procedures has limitations.
For example, although bronchoscopy has a sensitivity of
85% for central tumors, it is much lower for small and
peripheral lesions. The false-negative rate for transthoracic
 Figure 10-13 Patterns of bone marrow FDG uptake. A, Diffuse uptake in the marrow is
frequently seen in cancer patients following colony stimulating factor therapy. B, Osseous
metastases usually present with a heterogeneous pattern.

Figure 10-14 FDG uptake in the thymus. Area of uptake above the heart (arrow) in the typical
configuration of the thymus (A) corresponds to a normal thymus (B) on CT. After chemotherapy,
this uptake may be even more intense, the so-called thymic “rebound.”
 Oncology: Positron Emission Tomography 317

Figure 10-15 Metal artifact.Axial (A) and coronal (B) CT and PET images show beam hardening
artifact on the CT (left) from bilateral hip prostheses.The attenuation corrected PET images
(middle) show artifactually increased uptake along the lateral margins of the prostheses, which is
significantly decreased on nonattenuation-corrected images (right).

needle biopsy is approximately 25%. Transthoracic nee-
dle biopsy also carries a 10–25% risk of a pneumothorax
requiring a chest tube. Patients may require thoraco-
tomy and surgical biopsy for definitive diagnosis.
Diagnosis of Solitary Pulmonary Nodule
The use of F-18 FDG PET for pulmonary nodule evalua-
tion was one of the first indications in which PET was
proven to be clinically useful. A pulmonary nodule is
defined as a well-circumscribed lesion measuring <4 cm
in size. With the increased use of CT, the incidental
detection of these nodules has risen tremendously; about
one half will prove to be malignant.
The presence of central calcifications in some nodules
indicates they are benign granulomas. However, most
pulmonary nodules are indeterminate based on radio-
graphic appearance. The conventional workup of an
indeterminate nodule is biopsy or serial radiographic fol-
low-up for 2 years. If a nodule is stable in size over
a 2-year period,it is presumed benign. Suspicious-appear-
ing nodules and those that seem to increase in size are
sent to biopsy. This method results in biopsy of many
benign lesions and delayed diagnosis of some malignant
cases. F-18 FDG PET has proven to be an accurate
method to differentiate benign from malignant nodules
and decrease biopsy of benign lesions (Fig. 10-19A).
318 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-16 Respiratory motion artifact. Coronal (A) and axial (B) CT scans in lung windows
and corresponding PET scans show intense FDG activity in liver metastases as well as in the right
lung base. However, no lung mass is seen on CT. C, Two axial enhanced CT images show liver
metastases. Differences in respiration have caused misregistration and a posterior liver lesion
projects over the lung on PET.

Figure 10-17 Effects of arm position and beam hardening on
CT images. Soft tissue window CT image with the patient’s arms
up (A) compared with arms down (B) demonstrates the effect of
beam hardening with artifact throughout the abdomen on B.
Malignant nodules generally have increased F-18 FDG
uptake with an SUV >2.5, although most cancers have
considerably higher SUVs ( Fig. 10-19B). The sensitivity
of PET is reported to be about 95% and the specificity
approximately 81%. The high negative predictive value
of PET means that biopsy can usually be avoided when
the PET scan is normal. Because there is still a chance
Figure 10-18 Central tumor necrosis on PET. A left upper lung carcinoma seen as a solid mass
on CT actually contains significant central necrosis that is revealed as absent uptake on PET.
Visualization of regional differences in tumor metabolic activity with PET can help direct a biopsy.
Oncology: Positron Emission Tomography 319
that malignancy is present, these patients should be fol-
lowed with CT. The incidence of malignancy in patients
with a negative PET scan depends on the prevalence of
disease; it may be as low as 1% for patients at low risk for
cancer, but can be 10% in high-risk cases.
PET scans can change the surgical approach for nod-
ules demonstrating increased uptake. This includes
identification of mediastinal lymph nodes with abnormal
uptake and distant metastases ( Figs. 10-20 and 10-21).
In these patients, thoracotomy and surgical biopsy may
be changed to mediastinoscopy.
However, PET has some limitations. False-negative
results can be seen in small lesions <1 cm. The resolu-
tion of PET is on the order of 7–8 mm, and volume aver-
aging of small tumors with surrounding normal tissue
may result in low- or normal-appearing uptake. The
greater motion occurring with tachypnea or in the more
mobile lung bases may accentuate volume averaging.
False-negative PET results can also occur in certain
tumors. In the lungs, the most common of these are
bronchoalveolar cell carcinoma and carcinoid.
The positive predictive value of PET is not as high as
the negative predictive value, and a positive result cannot
be assumed to represent malignancy. An inflammatory
process such as granuloma commonly causes false posi-
tive findings. In some areas of the country where granu-
lomatous disease is highly prevalent, this is a very
significant problem.
A malignant solitary nodule is most commonly caused
by primary bronchogenic carcinoma, although a single
metastatic lesion is also a possibility. Close examination of
the PET scan and correlative CT are essential to detect any
unexpected primary tumor outside of the lungs.
Staging NSCLC
Staging of NSCLC lung carcinoma is critical in assessing
prognoses and deciding the appropriate course of therapy.
320 NUCLEAR MEDICINE: THE REQUISITES
Box 10-4 Factors Affecting Standard
Uptake Value (SUV) Levels
Factor Change in SUV
↑ Serum glucose ↓
↑ Body mass ↑
↓ Dose delivery: extravasated dose ↓
↑ Uptake period ↑
↓ Region-of-interest size ↑
↓ Pixel size ↑ NON-SMALL CELL
Box 10-5 Tumors with Low F-18 FDG PET
Uptake
Prostate cancer
Renal cell carcinoma
Bronchoalveolar cell lung cancer
Mucinous adenocarcinomas
Carcinoid
Low-grade sarcomas
Low-grade lymphoma
MALT mucosa-associated lymphoma
Small cell lymphocytic non-Hodgkin lymphoma
Differentiated thyroid cancer (iodine-avid)
Hepatocellular carcinoma
Metastasis to brain
Lung cancer staging is based on the tumor-node-metastasis
( TNM ) classification ( Boxes 10-7 and 10-8). Generally,
stage I and stage II patients are considered resectable,
although a subset of stage III patients might benefit from
surgery. Only about 25% of patients present with stage I or
stage II disease. However, improved methods of staging
are needed, as up to 50% of patients with NSCLC who
undergo curative surgery suffer a recurrence.
CT and PET often have complementary roles in the stag-
ing, restaging, and surveillance of NSCLC. CT better
assesses tumor size, invasion of the pleura and medi-
astinum, and the distance of the tumor from the carina.
Also, abnormalities such as atelectasis and aspiration pneu-
monia can cause abnormal F-18 FDG uptake that may be
confused with the primary tumor.
However, radiographic staging has its limitations. For
example, CT examination of the lymph nodes depends on
size criteria to determine if tumor involvement is present.
Any lymph node larger than 1 cm is considered abnormal
and suspicious for tumor involvement. This frequently
leads to understaging patients. PET can often detect
abnormal activity from tumor in normal-sized lymph
nodes. In addition, a whole-body PET scan may detect
Box 10-6 World Health Organization
Histologic Classification of
Lung Carcinoma
SMALL CELL
Pure small cell (oat cell)
Mixed (small cell and large cell)
Combined (small cell and squamous cell or
adenocarcinoma)
Large cell
Undifferentiated large cell
Giant cell
Clear cell
Squamous cell carcinoma
Epidermoid
Spindle cell
Adenocarcinoma
Acinar
Papillary
Bronchoalveolar
Solid carcinoma with mucus production
Adenosquamous carcinoma
Bronchial gland carcinoma
Adenoid cystic carcinoma
Mucoepidermoid tumor
Carcinoid
lesions not seen on a chest and abdomen CT examination
(Fig. 10-22). The sensitivity of PET for tumor involvement
in any one lymph node is 75%,but it averages 91% for over-
all mediastinal involvement. This compares well to
reported sensitivities of CT at 63–76%.
Both CT and PET can have false-positive results in the
lymph nodes. Inflammatory conditions can cause
enlarged lymph nodes on CT, with false positive interpreta-
tions occurring in up to 40% of patients. Inflammatory
processes and the effects of therapy also cause increased
uptake of radiotracer. However, lymph nodes may remain
enlarged after disease has resolved and PET has normal-
ized. Following therapy, PET provides information on
response that CT cannot. Patients who respond to therapy
will show PET changes more quickly than on CT.
The results of PET can help direct the method and loca-
tion of biopsy. Patients with no mediastinal involvement or
distant metastases can go to thoracotomy for curative
resection at the time of biopsy. C, However, PET may not
be able to differentiate N1 disease from N2 disease. This is
a critical difference, as N1 involvement is operable but N2
is not. Mediastinoscopy may be needed to differentiate N1
from N2 disease or to further evaluate the mediastinum
 Oncology: Positron Emission Tomography 321

Figure 10-19 Characterization of solitary pulmonary nodules. A, A left lung nodule on CT had
no FDG uptake on PET consistent with a benign process.This lesion remained stable on CT follow-
up confirming this impression. B, A small, well-circumscribed right lower-lobe nodule with
increased FDG accumulation on PET was later found to be an adenocarcinoma.

Figure 10-20 Detection of distant disease with PET.A large right-lung nodule showed markedly
increased FDG uptake consistent with tumor. Contralateral hilar lymph nodes were also abnormal,
which meant this patient was not a candidate for surgical resection.

when PET is positive. It is important to describe lymph
node involvement in uniform terms. The commonly used
classification system is outlined in Box 10-9 and Figures
10-23,10-24,and 10-C [see color insert].
PET is superior to conventional imaging modalities for
the detection of distant metastases. Bone involvement is
detected with a somewhat higher sensitivity than the
bone scan, and a considerably higher specificity. Tumor
involvement that might be overlooked on CT may be
seen with PET. This includes retroperitoneal and supra-
clavicular lymph nodes,soft tissue lesions,and small adre-
nal metastases. Assessment of the adrenal glands is
important, as they are a frequent site of lung cancer
metastasis ( Fig. 10-25). CT reveals adrenal lesions in
roughly 20% of cases, but the majority are later proven to
be benign adenomas. PET can help differentiate benign
 Figure 10-21 PET improves lung cancer staging. A, The right upper lobe bilobed pulmonary nodule
appears malignant on PET. B, Small lymph nodes on CT would not be read as positive based on size
criteria alone. C, However, PET revealed these small lower paratracheal lymph nodes to be abnormal.
Biopsy confirmed malignancy in both regions.

Box 10-7 Tumor-Node-Metastasis (TNM) and malignant adrenal gland lesions with a high degree of
Staging of Lung Carcinoma accuracy based on the level of uptake.
Overall,PET has a significant effect on patient staging and
PRIMARY TUMOR (T) management.Unsuspected metastases are detected 10–14%
Tx: Malignant cells; primary not seen of the time. A significant change in management has
T0: No evidence of primary tumor been reported in 19–41% of cases based on PET findings.
T1: <3 cm; surrounded lung or visceral pleura; not in
mainstem bronchus Restaging NSCLC and Assessing Response
T2:Any of the following: to Therapy
>3 cm When surgery and therapy distort anatomy, PET can
Involves mainstem bronchus >2 cm from carina detect residual and recurrent tumor not found on CT.
Invades visceral pleura
Associated atelectasis or obstructive pneumonia
extends to hilar region but does not involve whole
lung Box 10-8 Staging of Lung Carcinoma
T3:Any size invades chest wall, pericardium, mediastinal Based on Tumor-Node-
pleura, diaphragm Metastasis (TNM) Classification
<2 cm carina Scheme
Atelectasis entire lung

REGIONAL LYMPH NODES (N) Stage TNM

N0: No involvement Ia T1 N0 M0
N1: Nodes within lung, ipsilateral bronchopulmonary or Ib T2 N0 M0
hilar IIa T1 N1 M0
N2: Ipsilateral mediastinal, mid-line prevascular, and/or IIb T2 N1 M0
subcarinal T3 N0 M0
N3: Contralateral mediastinal, contralateral hilar, IIIa T3 N1 M0
contralateral scalene, supraclavicular T1–3 N2 M0
IIIb T4 N0–2 M0
DISTANT METASTASIS (M)
T1–4 N3 M0
M0: No distant metastasis IV Any T, any N, M1
M1: Distant metastasis
Based on the AJCC staging system, 1997.
 Oncology: Positron Emission Tomography 323

Box 10-9 Classification of Lymph Node

Stations in the Chest

Station Designation

SUPERIOR MEDIASTINAL NODES

1 Highest mediastinal Above top of left brachio-
cephalic vein crossing
trachea
2 Upper paratracheal Below station 1 to top aortic
arch
3 Prevascular Anterior to great vessels
above top aortic arch
Retrotracheal Behind trachea, thoracic inlet
to bottom azygous vein
4 Lower paratracheal Below top of aortic arch
Above top right upper lobe
bronchus
Azygous arch divides
superior from inferior

AORTIC NODES
5 Subaortic or AP Medial first branch
window pulmonary artery, lateral
ligamentum arteriosum
6 Para-aortic Anterior and lateral to aortic
arch

INFERIOR MEDIASTINAL NODES

Figure 10-22 Added value of whole-body scanning. The MIP 7 Inferior mediastinal Subcarinal
image of a patient with a left upper lobe non-small carcinoma 8 Paraesophageal Lateral to esophagus or
(above the heart) demonstrates a metastasis to an external iliac
anterior if below
node in the right pelvis, not detected previously as CT scans were
only done through the adrenal glands.
subcarinal nodes
9 Pulmonary ligament In pulmonary ligament

N1 NODES—DISTAL TO MEDIASTINAL PLEURAL

REFLECTION
Caution must be used as increased F-18 FDG accumula- 10 Hilar
tion may occur following therapy. The most significant 11* Interlobar Between lobar bronchi and
problem occurs after radiation. The accuracy of PET adjacent to lobar bronchi
can be improved by delaying imaging until at least 12* Lobar
3 months after therapy. Even given these limitations, 13* Segmental
PET can provide valuable information by identifying 14* Subsegmental
residual disease or relapse. For example, distant recur- *Intrapulmonary.
rence after complete resection of tumor occurs over
20% of the time. PET restaging frequently leads to
changes in management.
The use of PET to measure a response to treatment Small Cell Lung Carcinoma
is more limited. Although a decrease in the size of SCLC staging usually involves categorizing the disease as
a mass on CT has been used to assess response, this limited or extensive. If disease that is confined to one
does not always provide the best indicator of the effi- hemithorax, it can be treated more successfully by adding
cacy of therapy. A response to therapy with PET has radiation to the chemotherapy. Small cell lung carcinoma
been defined as a decrease in uptake on the SUV by at shows intense F-18 FDG accumulation. In general, data
least 50%. Limited data suggests a decrease in F-18 on the use of PET in SCLC is more limited than for NSCLC.
FDG activity may provide a better measure of response PET may help detect additional metastasis and lead to
and normalization of uptake may be associated with upstaging of some patients. The cost effectiveness of
a favorable prognosis. F-18 FDG PET in SCLC has not been proven.
324 NUCLEAR MEDICINE: THE REQUISITES
Figure 10-23 A–B, Lymph node classification of the chest. Coronal images of the chest outline
the location of lymph node stations described in Box 10-9.
Head and Neck Carcinoma
The prognosis of head and neck cancer depends on the
disease stage. As distant metastases at initial diagnosis
occur only 5% of the time, local lymph node status is
most critical in determining which patients are candi-
dates for surgical resection. At diagnosis, roughly 40% of
patients have localized disease and 60% have advanced
cases ( Figs. 10-26 and 10-27 ). F-18 FDG PET has been
found to be useful in staging, monitoring therapy
response, and detecting recurrence of head and neck
cancers ( Box 10-10). PET reportedly changes patient
management in 20–30% of patients with head and neck
cancer.
Although FDG PET is sensitive, it plays a more limited
role in head and neck tumor diagnosis than lung cancer.
Conventional modalities adequately visualize most
tumors and are generally better able to assess tumor size.
However, head and neck cancer often presents with pal-
pable adenopathy. These lymph node metastases may be
much larger than the primary tumor in head and neck
Continued
cancer. In 5% of cases, the primary cannot be identified
by endoscopy, CT, or MRI, which means that a patient
might have to undergo large-field radiation therapy.
Although small primary tumors may not be detected, PET
can identify the unknown primary tumor in 20–50% of
patients.
In the staging of head and neck cancer, PET has been
found equivalent if not superior to CT and MRI in the
detection of nodal metastases. When lymph nodes are
normal in size, this is particularly helpful. For restaging,
PET appears superior to conventional imaging modali-
ties. When the anatomy is distorted from surgery and
radiation, restaging and detection of recurrent tumor by
PET is superior to CT. The reported sensitivity of PET for
recurrence ranges from 79–96%, with a negative predic-
tive value >90%. CT, on the other hand, has a sensitivity
of 54–61%.
PET-CT is particularly useful in evaluating the post-
operative neck ( Figs. 10-28 and 10-D [see color insert]).
With a loss of symmetry, evaluation is difficult; fusion
images allow better identification of increased uptake in

Figure 10-23 B, cont’d.
normal structures, such as discriminating muscles, from
sites of tumor recurrence or metastases. Recent studies
suggest additional imaging after a further delay of an
hour or so may increase specificity.
Consistent terms must be used to describe the loca-
tion of head and neck cancer. Different methods of
describing the location of cervical lymph nodes have
been utilized over the years. Currently, an imaging-based
method of lymph node classification proposed by Som
optimizes recent updates by the AJCC ( Box 10-11; Figs.
10-29 and 10-30).
Thyroid Carcinoma
Thyroid cancer must be considered separately from
other cancers occurring in the head and neck. In most
cases, thyroid cancer derives from the follicular cells of
the gland giving rise to papillary, follicular, or mixed cel-
lularity variants. These differentiated tumors accumu-
late iodine-131 (I-131) and are best evaluated and treated
with radioactive iodine. The sensitivity of F-18 FDG in
these patients is low.
Oncology: Positron Emission Tomography 325
Clinical utility of FDG PET scanning is limited to thy-
roid malignancies that do not accumulate I-131. These
include poorly differentiated, aggressive tumors. This
may occur in metastatic and recurrent tumors that
degenerate from a previously well-differentiated,
iodine-avid tumor. In cases where the I-131 scan is
negative but serum thyroglobulin levels remain ele-
vated, the sensitivity of PET is >90%. PET may help
direct surgical resection or external beam radiation
therapy. Unlike I-131 scanning, patients do not need to
undergo thyroid hormone replacement therapy with-
drawal or stimulation with recombinant TSH. PET may
also be of some benefit in the more aggressive Hürthle
cell variant of follicular carcinoma and in some anaplas-
tic tumors.
PET scanning may also be useful in medullary thy-
roid carcinoma. This tumor arises from the parafollicu-
lar cells of the thyroid and does not accumulate I-131.
These tumors have been evaluated with somatostatin
receptor analogs, pentavalent DMSA, and thalium-201
and Tc-99m sestamibi. However, FDG PET may be
superior to these agents and can be helpful in difficult
326 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-24 Transaxial diagram chest lymph node station positions as outlined in Box 10-9
(pulmonary ligament [level 9] not included) and Figure 10-23.

cases. The sensitivity of PET has been reported to be
78% with a specificity of 79%.
It must be noted that F-18 FDG may accumulate with
equal intensity in benign adenomas, thyroiditis, and
malignant lesions. Although an incidentally detected
F-18 FDG avid nodule should be pursued to exclude
malignancy, PET has no role in the diagnosis of thyroid
cancer.
Esophageal Carcinoma
Esophageal carcinoma is most commonly due to squa-
mous cell cancer in the upper two-thirds of the esopha-
gus, whereas adenocarcinoma typically occurs in the
lower third. It frequently presents as dysphagia or is
detected by endoscopic biopsy in patients with Barrett’s
esophagus, a known precursor of many cases of
esophageal cancer. Because whole-body scanners have
limited sensitivity and specificity for detecting tumor in
patients with Barrett’s, PET has not proven useful in
screening these patients.
Diagnosis
Overall, the sensitivity of PET is 95% in esophageal can-
cer. The diagnosis of the primary esophageal tumors by
PET may be limited by a small tumor volume. In adeno-
carcinoma, 10–15% of patients may be falsely negative by
PET due to low uptake in mucinous and signet cell vari-
eties. PET is not able to determine the extent of the pri-
mary tumor and may not offer any substantial advantage
over the standard diagnostic modalities such as endo-
scopic ultrasonography.
 Oncology: Positron Emission Tomography 327

Figure 10-25 Adrenal metastases on PET. A, A non-small cell carcinoma of the right lung showed
mediastinal involvement on PET. B, The enlarged adrenal gland seen on CT also appeared malignant
by PET.Adrenal metastases can also be detected with PET in normal appearing glands by CT.

Figure 10-26 Recurrent squamous cell carcinoma of the head and neck. Coronal and sagittal
PET images demonstrate tumor activity in enlarged palpable left cervical lymph nodes, as well as the
primary tumor posteriorly along the oropharynx extending up to the skull base.
328 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-27 Head and neck carcinoma staging. PET images reveal several abnormal lymph
nodes in the right cervical and supraclavicular region, as well as an unexpected mediastinal
metastasis to a normal-sized lower paratracheal (arrow).

example, cervical metastases are more common in proxi-

Box 10-10 F-18 FDG Imaging in Head and mal tumors and abdominal lymph node involvement may
Neck Carcinoma be more frequent in distal masses. However, spread is
unpredictable.
Indication Utility of FDG PET Although the sensitivity of PET for lymph node
involvement is not high ( >70%), the specificity is >90%
Surveillance Very useful
( Figs. 10-31 and 10-E [see color insert]). Identification
Restaging recurrence Very useful, especially
postoperative neck but
of paraesophageal nodes is limited by PET as they may
limited by radiation be inseparable from the primary mass and very small
change lesions may be below the resolution of the PET detec-
Response to therapy May be useful tion systems. However, the detection of distant metas-
Diagnosis in unknown Useful; positive only tases with F-18 FDG is superior to CT. The accuracy of
primary 25–30% PET has been reported to be 83%, compared to 68%
Staging with CT. PET may result in a change in patient manage-
Clinical N1–N3 neck Useful, detect distant ment in a significant number of cases, approximately
metastasis 20% of the time.
Clinical N0 neck Low yield

Restaging and Monitoring Response to Therapy

Staging
Esophageal cancer most commonly spreads to regional
lymph nodes. The location of these nodes depends to
a certain extent on the level of the primary tumor. For
PET has proven value in assessing patients during ther-
apy and following therapy for recurrence. A scan fol-
lowing neoadjuvant chemotherapy may better predict
patient survival after subsequent therapy than standard
imaging methods ( Fig. 10-32). The ability to detect
 Oncology: Positron Emission Tomography 329

Figure 10-28 Restaging head and neck cancer.PET offers significant advantages in patients with
postoperative change where landmarks are gone or distorted.Following modified radical neck
dissection,this patient shows considerable asymmetry,but recurrent disease is obvious on the PET scan.

Box 10-11 Comparison of Nodal Classification Systems for Head and Neck Cancer

Rouviere System AJCC System Imaging-Based System

Submental I IA: medial to medial edge Below mylohyoid muscle, above hyoid
anterior belly digastrics bone
Submandibular I IB: lateral to IA and anterior
to back of submandibular
gland
Internal jugular II: skull base to hyoid, II: skull base to bottom of IIA: anterior, lateral, or inseparable
anterior to back edge hyoid, anterior to back from internal jugular vein
sternocleidomastoid edge sternocleidomastoid
IIB: posterior to internal jugular
vein with fat plane between
Retropharyngeal III: hyoid to cricothyroid III: bottom of hyoid to Lateral to carotid, level VI nodes are
membrane, anterior to bottom of cricoid arch, medial to carotids
back edge sternocleido- anterior to back edge
mastoid sternocleidomastoid
Midjugular IV: cricothyroid membrane IV: bottom of cricoid arch Lateral to carotids, level VI nodes are
to clavicle, anterior to to top of manubrium, medial to carotids
back edge of sternocleido- anterior to back edge
mastoid sternocleidomastoid
Spinal accessory V: posterior to sternocleido- V: posterior to sternocleido- VA: skull base to bottom of cricoid
mastoid, anterior to mastoid, anterior to arch
trapezius, above clavicle trapezius, above clavicle
VB: bottom cricoid arch to
level clavicle
Anterior VI: below hyoid, above VI: below bottom of hyoid, Visceral nodes
compartment suprasternal notch, above top of manubrium,
between carotid sheaths medial to carotid arteries
Upper mediastinal VII: below suprasternal VII: below top manubrium Overlaps highest mediastinal nodes of
notch and above innominate, chest classification
between carotids
Supraclavicular Clavicles in field of view,
above and medial to ribs

All systems use facial, parotid, retropharyngeal, and occipital groups.

330 NUCLEAR MEDICINE: THE REQUISITES
Figure 10-29 Cervical lymph node levels according to the imaging based classification system
described in Box 10-11.
residual and recurrent disease is also significant.
However, caution must be taken when evaluating
patients immediately after therapy as an artifactual
increase in activity may be seen ( Fig. 10-33).
Normal physiologic activity in the esophagus may
confound interpretation. Similarly, intense F-18 FDG
uptake in normal stomach limits the usefulness of PET
in evaluating tumors of the stomach and gastro-
esophageal junction. However, when a gastric tumor is
F-18 FDG avid, PET scanning may be useful in monitor-
ing therapy.
Colorectal Carcinoma
Colon cancer is known to develop in colon polyps.
Dysplastic elements are found in approximately one
third of adenomatous polyps. A progression to invasive
cancer occurs slowly. The diagnosis of colorectal carci-
noma depends on direct visualization by colonoscopy
as well as imaging with CT and barium enema. Staging
with CT often identifies regional adenopathy and dis-
tant metastases. The staging of colorectal carcinoma is
important in determining therapy and prognosis
( Boxes 10-12 and 10-13).
The ability of PET to diagnose primary colorectal
tumors may be limited as small tumors and cancerous
polyps are frequently not detected, and the high levels of
F-18 FDG activity normally occurring in the bowel can
decrease sensitivity and specificity. However, even given
these limitations, PET detects primary colon carcinomas
>90% of the time. In comparison, CT has a sensitivity of
about 60%.
Although contrast-enhanced CT is the primary
modality for the staging of colorectal carcinoma, it often
 Oncology: Positron Emission Tomography 331

Figure 10-30 Transaxial diagram of cervical lymph node stations at the level of the floor of the
mouth and submandibular gland (S) (A), the hyoid bone ( H) (B), thyroid cartilage and cricoid
cartilage (C), and just above the clavicles (C ) (D) with a portion of the thyroid gland (Th) in view.
Note the appearance of the sternocleidomastoid muscle (SC ), which is a key landmark.

fails to identify nodal disease. F-18 FDG PET is superior
to CT for initial staging and the detection of recurrent
colon cancer. Local recurrence can be difficult to
detect on CT due to scarrin following therapy, but may
be easily visible on PET images. The obturator and iliac
nodal regions of the pelvis are often particularly diffi-
cult to evaluate on CT, and PET is very helpful in evalu-
ating these regions. However, PET is also superior to
CT in the retroperitoneal nodes. The detection of
hepatic metastases is good with CT, but PET often com-
pliments this information or detects new lesions ( Fig.
10-34). Sensitivity of PET has been reported at 85–99%,
with specificity of 71–87%. PET may directly alter
patient management in 29–36% of staging and restaging
cases. Data are limited concerning the use of PET in
monitoring therapy effects.
Other Tumors of the Gastrointestinal Tract
The use of PET in other tumors of the gastrointestinal
tract is more limited, and CT remains critical in the
analysis of these tumors. However, PET may be help-
ful in tumors of pancreatic, biliary, and hepatic origin.
PET is often used in patients with CT scans that are dif-
ficult to evaluate or in patients with elevated serum
tumor markers. These include alpha-feto protein in
hepatocellular carcinoma and Ca 19-9 in pancreatic
cancer.
PET is highly sensitive for the detection of pancre-
atic cancer ( Fig. 10-35). However, CT is essential in
defining tumor extent and determining resectability.
Some masses seen on CT are actually benign, and PET
can often differentiate benign and malignant
processes ( with accuracies ranging from 85–93%).
This can support a negative fine-needle biopsy find-
ing. PET may also detect occult cancers not seen on
CT in symptomatic patients. The detection of nodal
disease is comparable between PET and CT. However,
PET may identify lymph nodes difficult to visualize on
CT, such as in the upper portal regions. Also, PET can
identify undiagnosed distant metastases in 14% of
cases. These factors can lead to alterations in surgical
332 NUCLEAR MEDICINE: THE REQUISITES
Figure 10-31 Identification of regional lymph nodes in head and neck cancer. A, Small lymph
nodes along the lesser curvature of the stomach (arrow) are seen on CT. B, Only low-level FDG
activity was present despite the presence of metastases on endoscopic biopsy.The difficulty in
identifying regional nodes may relate to activity in adjacent tumor or microscopic amounts of tumor
present.
Figure 10-32 Monitoring esophageal tumor response to
therapy. Sagittal PET scans (left ) before and (right) after
chemotherapy show rapid resolution of the abnormal activity
within an esophageal tumor. This type of response has been
linked to a better prognosis in recent studies.
management in up to 41% of cases. PET is often lim-
ited by poor sensitivity in small tumors and in acute
pancreatitis. The uptake in acute pancreatitis can be
as intense as in malignancy and can mask underlying
tumor. Acute pancreatitis often accompanies therapy
or obstruction by tumor, and PET in these cases is
often nondiagnostic.
Primary tumors of the liver are much less common
and are most often evaluated by CT. The two most com-
mon primary liver tumors are hepatocellular carcinoma
and cholangiocarcinoma. Although F-18 FDG PET is
highly accurate for detecting metastases to the liver, it is
considerably less sensitive for primary hepatic tumors.
PET may detect 50–70% of hepatocellular carcinomas,
but the sensitivity is reportedly higher ( 90%) for other
hepatic primary tumors.
Cholangiocarcinoma is a rare cancer of the bile ducts
that may not be detected on CT. It may occur in an extra-
hepatic or an intrahepatic location. Tumors arising
peripherally have a better prognosis as they may be
resected, whereas those near the hilum are infrequently
resectable. Tumors can be infiltrating, exophytic, or
a polypoid intraluminal mass. The sensitivity of PET is
lower for the infiltrating type. Gallbladder cancer is
a rare tumor that is generally diagnosed late in the course
of disease and may present with distant metastases. PET
may identify nodal metastases difficult to detect with CT,
including distant nodes and those high along the com-
mon biliary duct.
One tumor where PET has proven useful is the gas-
trointestinal stromal tumor (GIST ). These tumors usu-
ally show high levels of F-18 FDG accumulation. Rapid
response to Gleevac therapy seen with PET can predict
improved patient survival ( Fig. 10-36 ).
Lymphoma
Malignant lymphoma is classified as either Hodgkin dis-
ease (HD) (15%) or the more common non-Hodgkin lym-
phoma ( NHL) (85%). For Hodgkin disease, 10-year
survival rates are 80–85% for the early stages and ~40%
for very advanced (Stage IV ) disease. Survival rates are
much lower for NHL, with a 60% 5-year survival for the
potentially curable tumors, which are the high-grade,
more aggressive forms of the disease.
 Oncology: Positron Emission Tomography 333

Figure 10-33 Effects of radiation therapy on PET interpretation. A, Initial sagittal CT and PET
images reveal abnormal uptake in the esophagus from tumor. B, Two months following radiation
therapy, diffusely increased activity is seen in an extensive region of thickened esophagus. Although
this was presumed secondary to therapy, underlying tumor could go undetected and further follow-
up was needed.

HD tends to spread in an orderly fashion in contiguous

Box 10-12 Tumor-node-metastasis (TNM) lymph node chains. Staging of HD is important in treat-
Classification of Colon and ment planning and assessing prognosis, but surgical stag-
Rectal Carcinoma ing has largely been replaced by imaging ( Box 10-14).
Prognosis is good for stage I and stage II disease, which
Regional can be treated with local radiation therapy. More
Primary Lymph Distant advanced disease requires the addition of chemotherapy.
Tumor Nodes Metastases NHL is often widely disseminated at the time of
diagnosis and is more frequently fatal. Prognosis is closely
Tx: cannot be Nx: cannot be Mx: cannot be
related to histopathological classification and tumor grade.
assessed assessed assessed
T0: no evidence N0: none M0: none High-grade and intermediate-grade tumors are treated
of tumor with chemotherapy and radiation therapy with the goal of
TIS: carcinoma N1: 1–3 positive M1: distant a cure. Patients with low-grade tumors typically have an
in situ nodes metastases
T1: invades N2: ≥4
submucosal Box 10-13 Dukes Staging of Colorectal
T2: invades
Carcinoma
muscularis N3: central
propria nodes
T3: through A T1,T2; N0; M0
propria B T3,T4; N0; M0
T4: invades other C T1–4; N1; M0
organs D Any T; Any N; M1
334 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-34 Evaluation of metastatic rectal carcinoma.Axial-enhanced CT and PET images

before (A) and after (B) chemotherapy show a decrease in activity in a malignant rectal mass. C,
Coronal images before ( left ) and after (right ) therapy in this patient show a decrease in hepatic
lesions, only one of which was ever detected by CT.
 Oncology: Positron Emission Tomography 335

Figure 10-35 PET imaging of the pancreas. A, A malignant pancreatic head mass seen on
contrast-enhanced CT had a high level of F-18 FDG uptake consistent with malignancy. However, the
inflammation that accompanies pancreatitis can also cause intense uptake. B, CT following biopsy
of a pancreatic mass showed inflammatory changes around the pancreatic tail.These were positive
with PET. Note the central “cold”area corresponding to a pancreatic duct dilated due to proximal
obstruction from the mass.

indolent course initially. However, although therapy may
prolong survival, low-grade tumors relapse and eventually
transform to aggressive and fatal tumors.
Imaging
Developments in imaging and therapy have led to signif-
icant improvements in the treatment of lymphoma. CT
remains a critical component of diagnosis and staging.
Gallium-67 ( Ga-67 ) has long been used to stage, restage,
and evaluate the response of lymphoma to therapy.
However, there are drawbacks to this radiotracer. Spatial
resolution is limited, even when imaging is performed
with SPECT. The exam often requires multiple sessions
over the course of several days. The sensitivity of Ga-67
is limited below the diaphragm and in low-grade tumors.
Ga-67 imaging is discussed further in Chapter 9.
Recently, F-18 FDG PET imaging was proven to be use-
ful in the assessment of lymphoma. PET offers several
significant advantages over Ga-67, including improved
sensitivity ( Fig. 10-37). Imaging is usually completed
within hours and not days. Resolution is much higher,
radiation dose is lower, and quantification is possible
with PET. Specificity may be improved with image
fusion to the CT.
Detection
PET is highly sensitive in the detection of HD and high-
grade NHL and usually shows high levels of radiotracer
uptake. Some low-grade NHL cases caused by follicular
NHL are also accurately evaluated. Other low-grade NHL
tumors such as small cell lymphocytic lymphoma and
mucosa-associated lymphoma tissue ( MALT ) have much
less uptake and are not visualized reliably with F-18
FDG PET. Diagnosis of lymphoma generally relies on
histopathologic characterization of tissue samples, and
PET has not played a significant role in diagnosis of lym-
phoma. PET might be useful in directing biopsy to the
most accessible site.
336 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-36 FDG PET in gastrointestinal stromal tumor (GIST). PET has proven useful in
monitoring the remarkable effects of Gleevac therapy on GIST tumors. A baseline study (left ) is
needed to confirm the tumor is FDG avid. Rapid improvement is often seen within days of therapy
(right).

Staging, Monitoring Therapy, and Restaging
Although CT remains the primary modality for staging
lymphoma, the accuracy of PET is 96% compared to
56% with CT. With sensitivity >91%, PET scans are 10%
more sensitive than CT. The stage of the patient may
Box 10-14 Ann Arbor Classification
of Lymphoma
I Single lymph node or lymphoid structure
IE I+: growth into adjacent tissue or extralymphatic
involvement (not liver)
II Involving ≥2 regions on the same side of the
diaphragm
IIE II+: extralymphatic involvement
III Disease on both sides of the diaphragm (IIIS:
splenic involvement)
IIIE III+: involvement extranodal tissue localized
IV Nonlocalized or disseminated disease
be altered based on the PET in anywhere from 10–40%
of cases. PET is more accurate in assessing extranodal
disease, including bone marrow and spleen. Focal
lesions are generally caused by tumor, whereas diffuse
uptake may be the result of therapy. As bone marrow
biopsy can miss disease, a combination of PET and
biopsy may provide the most accurate evaluation of the
marrow.
FDG PET can accurately evaluate the effectiveness of
therapy. Decreased radiotracer uptake is seen in
patients responding to therapy ( Figs. 10-38 and 10-39).
This can be evaluated after as little as one cycle of
chemotherapy or even after a few days of therapy.
Responders identified by PET had longer disease-free
remission periods or were cured, whereas patients with
residual disease ( or nonresponders) all relapsed or pro-
gressed. This indicates PET may help optimize therapy
in patients with lymphoma.
Like Ga-67, PET can assess for tumor viability in
a residual mass found on CT following therapy. PET is
 Oncology: Positron Emission Tomography 337

Figure 10-37 Improved sensitivity of FDG PET over gallium-67. PET shows a large right neck
mass as involvement in the left neck, spleen and abdomen from lymphoma (A) whereas a 10-mCi
Ga-67 scan at 96 hours fails to detect lesion outside of the right neck (B).

much more specific than CT ( 86% compared with 31%)
in these cases ( Fig. 10-40). PET is often used in the
restaging of patients as information compliments that
gained by CT scanning.
Melanoma
The incidence of malignant melanoma is increasing dra-
matically. Survival depends on the stage at the time of
diagnosis. The thickness of the primary lesion is the most
important prognostic factor, and this is graded according
to the Breslow classification. Prognosis is extremely poor
with nodal or distant metastases. Even after potentially
curative surgery is performed, patients frequently present
with metastatic disease because of early hematogenous
spread. Some patients would benefit from further sur-
gery or directed therapy if they were accurately restaged.
Diagnosis of these metastases is difficult by conven-
tional modalities alone, such as CT. Metastatic disease
may occur in unusual locations such as other cutaneous
and subcutaneous sites, spleen, distant nodes, liver, and
gallbladder. Metastases are frequently found on PET in an
extensive pattern or widespread from the primary tumor
location ( Fig. 10-41). Thus, many sites perform head-to-
toe imaging on patients with melanoma. Lesions that are
not detected are likely microscopic. PET is generally more
sensitive than conventional imaging methods. However,
CT is more sensitive than PET in detecting small parenchy-
mal lung lesions and MRI best identifies brain metastases.
The sensitivity of PET is reported to be over 90% with
a specificity of 87%. PET frequently alters therapy in
20–26% of patients. PET is useful in staging patients at
high risk for metastases and in those who relapse.
PET does not replace sentinel lymph node scintigraphy
in patients with melanoma. Lymphoscintigraphy involves
Tc-99m sulfur colloid injection around the primary lesion
to identify the first draining sentinel lymph node.
Evaluating the resected sentinel lymph node with histo-
chemical staining is the most sensitive method to deter-
mine patients at risk for metastatic disease. This allows
detection of microscopic disease not detected with PET.
Breast Carcinoma
Breast cancer is classified as being a noninvasive or invasive
tumor of ductal or lobular type. Of invasive carcinomas,
338 NUCLEAR MEDICINE: THE REQUISITES

Figure 10-38 Restaging lymphoma. A, The initial PET image in a patient with non-Hodgkin’s
lymphoma shows extensive abdominal adenopathy, involvement of the spleen, chest and
supraclavicular nodes. B, Following two cycles of chemotherapy, much of the adenopathy has
resolved in the abdomen, but worsening disease is seen in spleen, bone marrow, and mediastinum,
requiring a change in therapy protocol.

ductal carcinoma accounts for 80% of cases, lobular 10%,
and medullary 5%. Noninvasive carcinoma, or carcinoma
in situ, may be detected by mammography when microcal-
cifications are present ( i.e., ductal carcinoma in situ
[DCIS]), but may be difficult to detect when it presents as
architectural distortion found in lobular carcinoma
in situ (LCIS).
Diagnosis
Mammography is a sensitive screening tool for detecting
breast carcinoma ( 81–90%). However, the specificity of
mammography is low and biopsy performed after an
abnormal mammogram results in a histopathological
diagnosis of malignancy only 10–50% of the time. MRI
continues to be investigated due to its high sensitivity
( up to 90–95%) for the detection of breast cancer,
although its specificity is also low. MRI is recommended
for patients with dense breasts, where the diagnostic
ability of mammography is limited. It can also better
visualize multifocal disease, recurrence, and disease in
patients with implants. Ultrasound has also added con-
siderably to the evaluation of the breast in cases of pal-
pable masses and discrete masses found on the
mammogram. Recently, F-18 FDG PET imaging has
proven to be useful in breast carcinoma staging and
restaging.
Primary Breast Cancer
The ability of F-18 FDG PET to detect primary breast
cancer is related to tumor size. One meta-analysis of
Figure 10-39 Monitoring therapy of lymphoma. A, PET-CT images show intense uptake in a
gastric lymphoma as well as an adjacent lymph node. Gastric involvement may not be detected, but
when seen, PET may be useful for follow-up. B, After one cycle of chemotherapy, no tumor could be
identified.This suggests a better prognosis than for a late or nonresponder.

Figure 10-40 Evaluation of a residual mass. A, During chemotherapy for lymphoma, a large
partially enhancing anterior mediastinal mass on CT showed markedly abnormal FDG accumulation.
B, When the follow-up CT showed residual mass, a follow-up PET was done. No FDG uptake was
seen consistent with fibrosis and scar.
340 NUCLEAR MEDICINE: THE REQUISITES
Figure 10-41 FDG PET in melanoma. A, Diffuse tumor
involvement including uptake near the primary tumor in the left
thigh, multiple lymph nodes, organs, and soft tissue metastases in
this patient with melanoma.This result led to changing planned
radiation therapy to systemic chemotherapy. PET can identify
subtle disease not found on CT. B, Regional involvement is seen in
metastases involving numerous right cervical lymph nodes in a
patient with a recently resected melanoma of the right ear.
the literature suggests the sensitivity of PET is 88% and
the specificity 79% for detecting primary tumors.
Reported sensitivity of PET is 92% for tumors measur-
ing 2–5 cm but is only 68% for tumors <2 cm. The
false-negative rate of F-18 FDG PET is not sufficient for
a screening test.
There may be a role for PET in the differentiation of
benign and malignant lesions. However, false negatives
may occur with well-differentiated and slow-growing
tumors, including lobular carcinoma, tubular carcinoma,
and DCIS. Lobular carcinoma is more difficult to detect
than invasive ductal carcinoma by mammography as
well. PET scanning may help detect multicentric dis-
ease. Also, PET may occasionally detect breast carci-
noma during evaluation of other malignancies.
One factor limiting the sensitivity of FDG PET is the
inherent resolution of whole-body PET scanners with
lesion detectability on the order of 6–8 mm. Dedicated
PET breast imaging systems are being developed, which
may significantly improve detection rates compared to
whole-body scanners.
Lymph Node Evaluation
Lymph node involvement has important prognostic and
therapeutic implications in breast carcinoma. Axillary
lymph node dissection fully evaluates the draining lymph
nodes. However, this is a highly invasive procedure with
side effects (e.g., lymphedema). In patients with nonpal-
pable lymph nodes, sentinel lymph node localization
with Tc-99m sulfur colloid, with or without blue dye, is
still the best method to select which patients should
undergo this highly invasive procedure. A negative PET
scan does not exclude the need for further workup.
F-18 FDG PET imaging is able to visualize lymph node
metastases by detecting changes in metabolism, often
before any anatomical change occurs on CT. Not all
lymph nodes will be visualized because microscopic
metastases will not be seen,and the resolution of PET pre-
vents evaluating the number of nodes involved.
Reported sensitivities of PET range from 79–100% and
specificities range from 66–100%. The accuracy depends
on the size of the primary tumor and the approach to
reading the exam. Positive lymph nodes are found more
often in larger tumors >2 cm with a sensitivity of 94%,
whereas metastases in small tumors may only be detected
one third of the time. If scans are interpreted in a highly
sensitive mode, a sensitivity of 95% is seen, resulting in
a high negative predictive value of 95%. However, this
method of reading lowers specificity to 66% because
inflammatory conditions frequently affect the axillary
lymph nodes and cause increased FDG accumulation.
Clinical Use of PET
Because of the limitations of F-18 FDG PET, it is not gen-
erally recommended in the initial diagnosis or screening
of most patients. However, it was proven to be useful for
staging patients with advanced disease, monitoring
response to therapy, restaging, and detecting recurrent
disease. Although PET cannot detect every metastatic
deposit, it frequently adds significant information to that
obtained with conventional imaging such as CT, MRI, and
bone scan. One study reported that PET added informa-
tion in 29% of cases (Fig. 10-42).
PET better detects distant metastases than conven-
tional modalities in chest, liver, and bone. In the bone,
PET is best able to identify more aggressive or lytic
 Oncology: Positron Emission Tomography 341

Figure 10-42 Breast cancer staging. A, PET often identifies malignant adenopathy in advanced and
recurrent breast cancer. In this case, the large right breast tumor shows markedly increased uptake, as
does an axillary lymph node. B, Abnormal breast uptake may not be due to primary tumor. Increased
uptake can be caused by therapy and hormonal stimulation.This postpartum patient presented with
lymphoma in the left cervical region and intense benign breast uptake.

lesions while the technetium bone scan visualizes scle-
rotic disease. Therefore, PET should not replace bone
scan in patients at highest risk for metastases. When
abnormalities are detected by CT, PET is often able to dif-
ferentiate benign from malignant processes.
More accurate assessment of tumor response to ther-
apy is possible with PET. Although CT can be used to
monitor therapy, it can only examine changes in size.
PET reveals changes in metabolic activity, which is
a more accurate indication of response. Limited data
suggests a 55% decrease in activity was able to accurately
differentiate responders from nonresponders after only
one cycle of chemotherapy. This allows rapid protocol
modifications to optimize therapy.
The detection of local recurrence is often difficult due
to distortion of the anatomy from surgery and radiation.
Yet detection is critical as recurrence occurs in up to
30% of patients. PET has a sensitivity of 90% for the
detection of recurrence ( Fig. 10-43). Caution must be
exercised because false-positive results can occur from
therapy, particularly radiation.
Ovarian Carcinoma
Tumors may arise from any of the cellular elements of
the ovary ( Box 10-15). Ovarian carcinoma diagnosis is
challenging because physical examination may not
reveal disease and symptoms do not present until late
in the course of disease. Tumor staging is outlined in
Box 10-16. Hematogenous spread is rare, but direct
spread and seeding of the omentum and organ surfaces
is common ( Figs. 10-44 and 10-F [see color insert]).
Lymphatic spread can lead to malignant pleural effu-
sions. Although patients with localized disease have
more than a 90% survival chance, most patients present
with stage III and IV disease. The prognosis of ovarian
carcinoma is poor, with overall survival of only 46% at
5 years.
342 NUCLEAR MEDICINE: THE REQUISITES

Preoperative evaluation of patients often includes

imaging with sonography, CT, and MRI. Staging with Box 10-15 Histopathologic Classification
CT has 70–90% accuracy. However, small peritoneal of Ovarian Carcinoma
lesions found with surgical exploration are frequently
overlooked or undetectable by CT. PET often high- EPITHELIAL TUMORS
lights abnormal activity in lesions that were over- Serous
looked on CT and this may be easiest to identify when Mucinous
the PET and CT are fused. Endometrioid
F-18 FDG PET has been used for staging and restaging Brenner
but is most widely used to detect recurrent disease. Transitional cell
Often,this involves patients with elevated serum markers Mixed epithelial
( Ca-125, Ca 19-9, alpha fetoprotien [AFP], and human Small cell
chorionic gonadotropin [HCG]) and negative or incon- Clear cell
clusive CT findings. The reported sensitivity of PET SEX CORD STROMAL TUMORS
varies from 50–90% and the specificity from 60–80%. Granulosa cell
The accuracy of PET depends on tumor size and cell Serotoli-Leydig cell
type. Small peritoneal nodules seen during laparoscopy
and small primary tumors confined to the ovary may be GERM CELL
missed. Well-differentiated and mucinous tumors may Dysgerminoma
not be seen, causing false negative results ( Box 10-17). Yolk sac
Also, PET scanning may not be useful for initial tumor Embryonal
diagnosis because several benign conditions may accu- Teratomas
mulate F-18 FDG (Box 10-18). Despite these limitations,
PET alters management in approximately 15% of cases.

Cervical Carcinoma
Cervical carcinoma is the most common gynecological
cancer. It may be treated effectively by surgery when
localized, but radiation and chemoradiation may be
required for locally advanced disease. The spread of
cervical carcinoma can occur by local extension or by
lymphatic spread to pelvic, para-aortic, retroperitoneal,
and even supraclavicular lymph nodes. Detection of
nodal involvement is important in therapy planning but
may be difficult by CT. In patients who have had prior
Box 10-16 Outline of Ovarian Carcinoma
Staging
Stage I: Growth limited to ovaries with an intact capsule
Stage II: Extension onto pelvic organs or into ascites
Stage III: Peritoneal implants outside of the pelvis;
retroperitoneal or inguinal lymph node involvement
Stage IV: Distant metastasis

Figure 10-43 Recurrent breast carcinoma in an internal mammary lymph node. PET can identify
metastases to regional lymph nodes as well as distant disease. Internal mammary node involvement
is frequent in cancers of the inner or medial breast.
 Oncology: Positron Emission Tomography 343

Figure 10-44 Recurrent ovarian carcinoma. A patient with a rising CA-125 had numerous
metastases on the coronal PET including a left axillary lymph node, metastases studding the surface
of the liver, and a right lower quadrant peritoneal lesion studding the right colon on axial images.

resection, scar may make detection of recurrent disease

by CT difficult. Box 10-17 Processes Causing False-
PET has shown >90% sensitivity for detection of cervi- Negative Results in Ovarian
cal cancer. Markedly abnormal F-18 FDG uptake is Cancer
seen in recurrent cancers and lymph node metastases
(Fig. 10-45). The use of PET requires careful correlation Well-differentiated serous cystadenocarcinoma
with CT. Evaluation may be complicated by normal Mucinous cystadenocarcinoma
uptake in bowel and urinary tract, as well as increased Disseminated peritoneal carcinomatosis
uptake in tissues affected by radiation therapy. Borderline tumors
Stage I tumors confined to the ovary

Testicular Carcinoma
Testicular cancer can be divided into seminoma and non-
seminoma germ-cell tumors. The spread of testicular
tumor is most commonly to retroperitoneal lymph nodes.
Although the overall prognosis for these tumors is good,
accurate staging and surveillance are necessary, but
patients are frequently incorrectly staged by CT. Patients
initially classified as stage I are commonly then found to
have nodal involvement at surgery. Other patients placed
incorrectly in high-risk groups may undergo unnecessary
therapy. For example, it has been common practice to
treat all patients with seminoma with radiation.
Standard imaging of testicular carcinoma consists of
ultrasound for diagnosis and CT for staging. PET is accu-
rate for tumor staging and detecting recurrence. PET has
a high negative predictive value of 94% and a sensitivity
344 NUCLEAR MEDICINE: THE REQUISITES

tumors. Although some have suggested that the urinary

Box 10-18 Processes Mimicking Ovarian
Carcinoma Causing False-
Positive F-18 FDG PET Results
Gastrointestinal activity
Infection/inflammation
Benign tumors:
Germ cell: benign teratomas
Epithelial tumors: mucinous cystadenoma, serous
cystadenoma
Dermoid cysts, hemorrhagic follicle cyst, corpus luteum
cyst
Endometrioma
Fibroma
Benign thecoma
Schwannoma
of 85%. Detection of small tumors and well-differenti-
ated teratomas is limited with PET. Relapse of testicular
carcinoma is a frequent occurrence and PET is useful for
surveillance.
Prostate Carcinoma
F-18 FDG PET has very limited sensitivity for prostate car-
cinoma. Uptake in primary tumor is low and similar as
for benign prostatic hypertrophy. In terms of staging,
F-18 FDG PET detected less than two thirds of osseous
metastases found on bone scintigraphy and approxi-
mately one half of nodal metastases. CT is superior to
PET for the detection of pulmonary metastases.
Renal and Bladder Carcinomas
CT is the most common imaging modality for the diagno-
sis and staging of renal cell carcinoma and bladder carci-
noma. PET is not useful in the diagnosis of primary
excretion of F-18 FDG might obscure adjacent tumors,
these masses often show no radiotracer uptake ( Fig.
10-46). The cause for this finding, such as variations in
glucose transporter expression, is being investigated.
However, F-18 FDG PET may play a role in diagnosing dis-
tant metastases and detecting recurrent disease.
There are limited data concerning the use of F-18 FDG
PET in bladder and renal tumors. Recent studies
detected only two thirds of primary bladder carcinomas,
but PET showed a high sensitivity for distant metastases.
In renal cell carcinoma, anywhere from 50–75% of pri-
mary tumors were identified by PET, although metastatic
lesions may be identified that are not seen on CT.
Musculoskeletal Tumors
Malignant primary bone tumors are usually F-18 FDG
avid,as are many benign conditions. Benign tumors such
as giant cell tumor, fibrous dysplasia, and eosinophilic
granulomas have been shown to accumulate F-18 FDG.
PET may be useful both in the evaluation of patients who
cannot undergo MRI imaging and in monitoring the
effects of therapy. If a nonresponder is identified early
by showing little change in SUV values on PET,the course
of therapy can be altered. F-18 FDG may be impact ther-
apy by identifying other sites of disease, such as in
patients with plasmacytoma.
For the evaluation of soft tissue sarcomas, the accu-
racy of F-18 FDG PET appears related to tumor grade.
The increased uptake in high-grade tumors such as malig-
nant fibrous histiocytoma allows detection with a high
degree of sensitivity. Low-grade tumors, on the other
hand, show minimal or nonexistent uptake, leading to
poor sensitivity. Although MRI remains the main imag-
ing modality of the primary tumor, the ability of MRI to
detect recurrence is limited in the postoperative patient.
PET may help detect recurrent tumors, although the
effects of surgery and radiation therapy lower sensitivity.

Figure 10-45 Cervical carcinoma. Residual tumor uptake is seen around air in the vaginal cuff
and increased metabolic activity is seen in a metastasis to the right pelvic sidewall.

Figure 10-46 Bladder carcinoma. Renal cell carcinoma and bladder tumors are often negative on
FDG PET.A mass along the posterior bladder (arrow) on CT ( left) shows no FDG uptake on the PET
portion of the exam (right).
SUGGESTED READING
Been LB, Suurmeijer AJ, Cobben DC, et al: F-18 FLT-PET in oncol-
ogy: current status and opportunities. Eur J Nucl Med 31:1659-
1672, 2004.
Beyer T, Antoch G, Muller S, et al:Acquisition protocol considera-
tions for combined PET/CT imaging. J Nucl Med 45:25S-35S,
2004.
Delbeke D, Martin WH: PET and PET-CT for the evaluation of
colorectal carcinoma. Semin Nucl Med 34:209-223, 2004.
Eubank WB, Mankoff DA: Evolving role of positron emission
tomography in breast cancer imaging. Semin Nucl Med
35:84-95, 2005.
Gupta NC, Maloof J, Gunel E: Probability of malignancy in soli-
tary pulmonary nodules using fluorine-18-FDG and PET. J Nucl
Med 37:943-947, 1996.
Ko JP, Drucker EA, Shepard JO, et al: CT depiction of regional
node stations for lung cancer staging. AJR 174:775-782, 2000.
Kresnik E, Mikosch P, Gallowitsch HJ, et al: Evaluation of head
and neck cancer with F-18 FDG PET: a comparison with con-
ventional methods. Eur J Nucl Med 28:816-821, 2001.
Kubota K, Yokoyama J, Yamaguchi K, et al: FDG-PET delayed
imaging for the detection of head and neck cancer recurrence
after radio-chemotherapy:comparison with MRI/CT. Eur J Nucl
Med 31:590-595, 2004.
Oncology: Positron Emission Tomography 345
Kumar R, Alavi A: PET imaging in gynecologic malignancies.
Radiol Clin North Am 42:1155-1167, 2004.
Macapinlac HA: FDG PET and PET/CT imaging in lymphoma
and melanoma. Cancer J 10:262-270, 2004.
Mountain CF, Dresler CM: Regional lymph node classification
for lung cancer staging. Chest 111:1718-1723, 1997.
Som PM, Curtin HD, Mancuso AA: Imaging-based nodal classifi-
cation for evaluation of neck metastatic adenopathy. AJR
174:837-844, 2000.
Vesselle H, Pugsley JM, Vallieres E, et al: The impact of fluo-
rodeoxyglucose F-18 positron-emission tomography on the sur-
gical staging of non-small cell lung cancer. J Thorac Cardiovasc
Surg 124:511-519, 2002.
Wahl RL: Principles and Practice of Positron Emission
Tomography. Baltimore, Lippincott Williams & Wilkins, 2002.
Wong RJ, Lin DT, Schöder H, et al: Diagnostic and prognostic
value of F-18 fluorodeoxyglucose positron emission tomogra-
phy for recurrent head and neck squamous cell carcinoma.
J Clin Oncol 20:4199-4208, 2002.
Zanzonico P: Positron emission tomography: a review of basic
principles, scanner design, and performance, and current sys-
tems. Semin Nucl Med: 87-111, 2004.
Figure 10-A Paralyzed vocal cord artifact on positron emission tomography (PET).Axial
computed tomography (CT), PET, and fused PET-CT images reveal unilateral uptake in the neck
localizing to the right vocal cord.The left vocal cord was found to be paralyzed on physical exam
after radiation therapy for lung cancer affected the left recurrent laryngeal nerve.This uptake could
be confused with a lymph node metastasis on PET if CT correlation is not used.

Figure 10-B FDG positron emission tomography (PET) appearance of “brown fat”activity. A,
Abnormal supraclavicular uptake localizes to the fat on fused PET-CT images.The fused images help
to differentiate this common benign variant from muscle activity or malignant adenopathy. B,
Lymphoma involvement in the supraclavicular region can appear similar to brown fat uptake on PET
images, but fused images localize the activity to involved lymph nodes.
 Figure 10-C Lymph node stations or levels of the mediastinum.

Figure 10-D Recurrent head and neck carcinoma. Recurrent or residual head and neck cancer is
often difficult to diagnose by computed tomography (CT) or magnetic resonance imaging due to
the distorted anatomy from surgery and radiation. Positron emission tomography (PET) better
demonstrates tumor even in cases of marked anatomical asymmetry. PET-CT makes localization
easier such as in this patient following removal of the sternocleidomastoid muscle and other
structures.
Figure 10-E Tumor involvement in regional lymph nodes. Small lymph nodes are seen in the
lesser curvature of the stomach on computed tomography. Positron emission tomography images
show only mild activity in one of the lymph nodes (arrow) near the tumor in the gastroesophageal
junction and fundus.This is better localized on the fused image (right). Endoscopic biopsy later
showed tumor involvement in the regional nodes during staging.

Figure 10-F Metastatic ovarian cancer along the right colon. Ovarian cancer spread often
presents as focal activity studding the peritoneal organs on positron emission tomography (PET).
Fused PET-CT images improve accuracy over PET alone or PET and CT read side-by-side. Uptake is
often subtle and can be confused with normal structures such as benign activity in bowel lumen.
 11
CHAPTER Gastrointestinal
System

Esophageal Motility Heterotopic Gastric Mucosa

Esophageal Motor Disorders Radiopharmaceutical
Achalasia Mechanism of Uptake
Scleroderma Dosimetry
Diffuse Esophageal Spasm Clinical Indications
Nutcracker Esophagus Meckel’s Diverticulum
Other Esophageal Disorders Clinical Manifestations
Scintigraphy Diagnosis
Radiopharmaceuticals Methodology
Dosimetry Gastrointestinal Duplications
Methodology Retained Gastric Antrum
Analysis and Quantification Barrett’s Esophagus
Accuracy Intestinal Function and Transit
Gastroesophageal Reflux Schilling Test
Diagnosis Intestinal Transit Scintigraphy
Scintigraphy Small Bowel Transit
Methodology Large Bowel Transit
Image Interpretation Protein-Losing Enteropathy
Accuracy
Gastric Motility Radionuclide studies of the gastrointestinal tract provide
Physiology noninvasive functional imaging and quantification of
Gastric Stasis Syndromes gastrointestinal function not available from other
Diabetic Gastroparesis methodologies.
Nonulcer Dyspepsia The gastric emptying study is the standard clinical test
Pharmacological Therapy for quantification solid and liquid motility. Gastro-
Scintigraphy esophageal reflux scintigraphy is an extremely sensitive
Clinical Indications and noninvasive method for diagnosis and quantification
Radiopharmaceuticals of reflux, used most commonly for pediatric patients.
Methodology Gastrointestinal bleeding scintigraphy has long been
Evaluation of Therapeutic Effectiveness used to localize the site of active bleeding prior to
Helicobacter Pylori Infection angiography. Tc-99m pertechnetate can localize
Urea Breath Test a Meckel’s diverticulum responsible for small-bowel
Gastrointestinal Bleeding bleeding or other heterotopic gastric mucosa.
Contrast Angiography Nonimaging radionuclide studies of gastrointestinal
Radionuclide Scintigraphy function date back to the earliest days of nuclear medicine.
Tc-99m Sulfur Colloid The Schilling test has long been the standard method for
Tc-99m Red Blood Cells the diagnosis of vitamin B12 malabsorption and pernicious
Tc-99m Red Blood Cells versus Tc-99m Sulfur Colloid anemia. Our new understanding of the infectious origin of

346
 Gastrointestinal System 347

gastritis and ulcer disease has led to a simple diagnostic The diagnosis of esophageal motility disorders is most
radionuclide C-14 breath test for detection of the bacteria often made by contrast radiography and esophageal
responsible, Helicobacter pylori. manometry. Although a barium-swallow study can
demonstrate anatomical lesions and mucosal changes, it
provides only a qualitative assessment of motility.
ESOPHAGEAL MOTILITY Esophageal manometry is the accepted reference stan-
dard for the diagnosis of motility disorders. It can quan-
The esophagus has multiple functions, including the tify peristaltic contraction, sphincter pressure, and upper
transport of food from the mouth to the stomach, clear- and lower esophageal sphincter (LES) relaxation, but it is
ance of regurgitated substances, and prevention of invasive and technically demanding. Although scintigra-
tracheobronchial aspiration and acid reflux ( Fig. 11-1). phy has been used as a diagnostic technique, it has been
The most common clinical complaint of patients with found most valuable in evaluating response to therapy.
abnormal esophageal motility is pain with swallowing
(dysphagia). Achalasia
This primary esophageal motor disorder manifests as par-
tial or absent relaxation of the lower esophageal sphinc-
Esophageal Motor Disorders ter and loss of esophageal body peristalsis, resulting in
Esophageal motor disorders have been classified as pri- esophageal retention of food and dilation, producing
mary (e.g., achalasia) and secondary (e.g., scleroderma) symptoms of dysphagia, weight loss, nocturnal regurgita-
or by the type of dysfunction (amotility, hypomotility, tion, cough, and aspiration. Etiology is unknown.
hypermotility) (Box 11-1). Upper gastrointestinal radiographic studies show reten-
tion of contrast in a distended column,narrowed sphincter,
and delayed clearance. Esophageal manometry reveals an
absence of peristalsis in the distal two thirds of the esopha-
mm WS
Hg
19 cm
gus, increased lower esophageal sphincter pressure, and
150
120 incomplete sphincter relaxation with swallowing.
Pharynx 90
60
30 Scleroderma
Upper
sphincter
0 Smooth muscle involvement of the esophagus is a com-
90 22 cm mon manifestation of this connective tissue disorder.
Upper 60
30
third
Aortic 0
arch 60 27 cm Box 11-1 Classification of Esophageal
30
Motility Disorders
0
Esophageal Middle
body third 60 32 cm
30 PRIMARY/SECONDARY
0 Primary
Lower 60 37 cm
Achalasia
third 30
0
Esophageal spasm
42 cm
Nutcracker esophagus
Lower 60
30
sphincter Secondary
0

Stomach 1 sec Scleroderma

Diabetic enteropathy
Figure 11-1 Esophageal anatomy and function. Swallowing
initiates a coordinated peristaltic contraction that propagates DEGREE OF MOTILITY
down the esophagus. The esophagus has three distinct regions: Amotility
the upper esophageal sphincter (UES), which allows food to pass
from the mouth to the esophagus and prevents tracheobronchial
Achalasia
aspiration; the esophageal body, with striated muscle proximally Scleroderma
and smooth muscle distally; and the lower esophageal sphincter
Hypomotility
(LES), a high-pressure smooth muscle region that prevents gastric
reflux but relaxes during swallowing to allow passage of food into Presbyesophagus
the stomach. Manometric pressure changes with a water swallow
(WS) of an 8-ml bolus. Following swallowing, the UES pressure falls Hypermotility
transiently. Then the LES pressure falls and remains low until Diffuse spasm
peristaltic contraction passes aborally through the UES and Nutcracker esophagus
esophageal body, which closes the LES.
348 NUCLEAR MEDICINE: THE REQUISITES

Contrast radiography demonstrates a dilated aperistaltic Radiopharmaceuticals

esophagus with barium retention and gastroesophageal
reflux. Manometry confirms decreased or absent lower
esophageal sphincter pressure and reduced contraction
amplitude.
Diffuse Esophageal Spasm
Symptoms include intermittent chest pain and dyspha-
gia. Barium swallow and manometry detect abnormal
nonperistaltic contractions of the esophageal body. No
organic lesion is demonstrable.
Nutcracker Esophagus
Patients present with noncardiac chest pain.Radiographic
studies are normal. The diagnosis is made with
esophageal manometric findings of high-amplitude, pro-
longed peristaltic contractions.
Other Esophageal Disorders
Smooth muscle disease of the esophagus also occurs
with systemic lupus erythematosus and polymyositis.
Esophageal striated muscle abnormalities are problems
for patients with muscular dystrophy, myasthenia gravis,
and myotonia dystrophica. Diabetes and alcoholism are
associated with abnormalities of esophageal motor
function. Severe esophagitis may also cause disordered
motility.
Esophageal transit scintigraphy is performed with
technetium ( Tc)-99m sulfur colloid (SC) or Tc-99m
diethylene-triamine-pentaacetic acid (DTPA), 200–300 μCi
(7.4–11.1 MBq), dispersed in a liquid bolus (usually
water). Reports suggest that semisolid food boluses are
more sensitive than liquid boluses for detecting dys-
motility. Transit is faster for less viscous materials, for
small versus larger volumes, and in the upright versus
the supine position.
Dosimetry
Table 11-1 provides the radiation absorbed dose (rad) for
gastroesophageal scintigraphy (esophageal transit, gas-
troesophageal reflux, and gastric emptying) for children
of different ages. Table 11-2 reports the dosimetry based
on the radiopharmaceutical and particular meal used.
The large intestine receives the highest radiation
absorbed dose.
Table 11-1 Dosimetry for Tc-99m SC Gastro-
esophageal Scintigraphy for Children
Rad/100 μCi by age*
Organ Newborn 1 year 5 years 10 years

Scintigraphy Stomach 0.383 0.093 0.050 0.031

Large intestine 0.927 0.380 0.194 0.120
Esophageal transit scintigraphy provides noninvasive Ovaries 0.099 0.042 0.033 0.072
functional diagnostic information about esophageal Testes 0.018 0.007 0.003 0.011
motility. The procedure is relatively simple to perform Whole body 0.020 0.011 0.006 0.004
and quantitative. It can be used to screen symptomatic
*Usual dose, 200–300 μCi (7.4–11.1 MBq).
patients and evaluate the effectiveness of therapy for Modified from Castronovo FP: Gastroesophageal scintiscanning in a pediatric
patients with diagnosed esophageal motility disorders. population: dosimetry. J Nucl Med 27:1212-1214, 1986.

Table 11-2 Adult Dosimetry for Esophageal and Gastric Scintigraphy

Millirads/study meal, by organ

Small Large Total

Stomach intestine intestine Ovaries Testes body

LIQUID
300 μCi Tc-99m SC 28 83 160 29 2 5
1 mCi Tc-99m DTPA 93 280 520 98 5 20
250 μCi In-111 DTPA 110 490 2000 420 27 60

SOLID
500 μCi Tc-99m SC ovalbumin 120 120 230 42 2 9
500 μCi Tc-99m chicken liver 120 120 230 42 2 9

Modified from Siegel JA,Wu RK, Knight C, et al: Radiation dose estimates for oral agents used in the upper gastrointestinal diseases. J Nucl Med 24:835-837, 1983.

Box 11-2 Esophageal Transit Scintigraphy:
Protocol Summary
PATIENT PREPARATION
Order an overnight fast.
Place radioactive marker on cricoid cartilage.
Position the patient supine.
Practice swallows with nonradioactive bolus.
RADIOPHARMACEUTICAL
Tc-99m SC or DTPA, 11 mBq (300 μCi) in 10 ml of water.
INSTRUMENTATION
Camera setup:Tc-99m photopeak with 20% window
Computer setup: 0.8-sec frame × 240; byte mode,
64 × 64
SWALLOWING PROCEDURE
Swallow Tc-99m SC or DTPA as a bolus.
Dry swallow at 30 sec, then radiolabeled bolus
every 30 sec × 4.
No swallowing between boluses.
PROCESSING
Time–activity curves, condensed dynamic images
QUANTIFICATION
Time to 90% emptying
Transit time
Methodology
Box 11-2 describes a standard liquid esophageal motil-
ity protocol and Box 11-3 describes a semisolid transit
protocol.
Although both anterior and posterior views can be
acquired, posterior acquisition allows for easier adminis-
tration of the bolus and closer observation of the patient.
The supine position is preferable because it eliminates
the effect of gravity on esophageal emptying. Because
gravity is the only mechanism of emptying in achalasia,
upright positioning is preferable for serial studies in this
disease. Emptying occurs in both the supine and the
upright position in systemic sclerosis.
Multiple swallows are often necessary for complete
emptying even in normal subjects because of a 25% inci-
dence of “aberrant” swallows, or extra swallows, which
occur between the two prescribed swallows. Aberrant
swallows inhibit subsequent swallows and delay transit.
Any normal residual remaining after an initial swallow
will clear when followed by a dry swallow.
Analysis and Quantification
Image analysis and computer cinematic movie display is
often adequate for diagnosis of severe motility abnormal-
Gastrointestinal System 349
Box 11-3 Esophageal Transit
Scintigraphy (Semisolid Meal):
Protocol Summary
PATIENT PREPARATION
Overnight fast.
Patient position: seated
Posterior camera acquisition
RADIOPHARMACEUTICAL
Tc-99m SC or DTPA , 37 MBq (1.0 mCi), mixed with
120 ml milk, 19 g of cornflakes and 1 g sugar
COMPUTER SETUP
Feeding phase: 10 sec frames × 12 (64 × 64 matrix)
Spontaneous emptying phase: 10 sec frames × 120
Water ingestion phase: 10 sec × 6
Carbonated beverage phase (150 ml): 10 sec × 6
PROCESSING
Draw a region of interest around the esophagus.
Display the time–activity curve
Display a condensed image
QUANTIFICATION
Calculate the percent retention = esophageal
counts/total counts × 100
Normal percent retention, <5% at 20 minutes
ities ( Figs. 11-2 and 11-3). However, quantitative analysis
is helpful for diagnosing less severe abnormalities, com-
paring serial studies, and evaluating the effectiveness of
therapy. Time–activity curves can be derived for the
entire esophagus or for the proximal, middle, and distal
thirds ( Figs. 11-3C and 11-4).
Esophageal transit can be quantified by calculating
either the transit time or the residual activity in the
esophagus. Transit time is the time from the initial
entry of the bolus into the esophagus until all but 10% of
peak activity clears. Abnormal for liquid bolus is >15
seconds (see Box 11-2).
Residual esophageal activity (%) = [(Emax – Et)/Emax] × 100
where Emax is the maximum counting rate in the esopha-
gus (15-second intervals) and Et is the counting rate
after dry swallow number t (abnormal, >20%). For
a semisolid meal, the abnormal retention at 20 minutes is
>5% ( see Box 11-3).
Functional images can be helpful for assimilating and
interpreting the many images acquired in a single transit
study. Because craniocaudal transit, not lateral motion, is
needed, the dynamic data can be condensed into a single
image with one spatial (vertical) and one temporal
dimension ( Figs. 11-5 and 11-6).
350 NUCLEAR MEDICINE: THE REQUISITES

Figure 11-2 Normal esophageal clear liquid transit. A, Images were obtained supine at 2-second
intervals. The swallowed bolus travels rapidly through the esophagus. Transit time is 11 seconds
(normal, 15 seconds). B, Scleroderma. Dysmotility with poor propagation of bolus and retention of
activity in the distal esophagus.

Accuracy of gastroesophageal reflux include esophagitis, bleeding,

Esophageal transit studies have a high sensitivity for the
diagnosis of achalasia but a generally lower detection rate
for other conditions, thus limiting their routine use as
screening tests. The specificity of scintigraphic findings
(e.g., for differentiating achalasia from scleroderma) is
also not high. Thus, in current practice, esophageal tran-
sit studies are most commonly done to evaluate response
to pharmacological, medical, or surgical therapy.
GASTROESOPHAGEAL REFLUX
Symptomatic reflux of gastric contents into the esopha-
gus is a common medical problem.Serious complications
perforation, Barrett’s esophagus, cancer, and stricture.
The clinical presentation of infants and children differs
considerably from that of adults. In adults, heartburn is
the usual complaint. Common pediatric symptoms
include respiratory symptoms, iron deficiency anemia,
and failure to thrive.
Gastroesophageal reflux occurs in infants as a normal
physiological event that resolves spontaneously by 7–8
months of age.Clinically important reflux becomes evident
by 2 months of age. Most children have a benign course
and are symptom free by 18 months of age. However,
approximately one-third have persistent symptoms until
age 4 years and 5–10% percent may have serious sequelae
of strictures,recurrent pneumonia,and inanition.

The adequacy of esophageal clearance is an important
factor in determining whether reflux becomes clinically
evident. Delayed clearance increases the duration of
mucosal exposure to refluxate. Other factors include the
efficacy of the antireflux mechanism, the volume of gas-
tric contents, the potency of refluxed material (acid,
pepsin),mucosal resistance to injury,and mucosal repara-
tive ability. Most adult patients with moderate to severe
esophagitis have a sliding hiatal hernia; however, the
majority of individuals with a hiatal hernia do not have
reflux disease.
Although lower esophageal sphincter pressure is
reduced in many patients with reflux, overlap exists
between healthy and ill subjects. Reflux events result
from either a transient sphincter relaxation not associ-
ated with swallowing, transient increases in intra-
abdominal pressure, or free reflux across an atonic
sphincter.
Diagnosis
Various tests have been used to diagnose gastroesoph-
ageal reflux. Barium esophagography can detect
mucosal damage,stricture,and tumor,but has a low sensi-
Figure 11-3 Achalasia: semisolid meal. A, Ten second frames × 12. Retention of activity in the
esophagus, most prominently in the distal portion.
Gastrointestinal System 351
tivity for detecting reflux and results in considerable radi-
ation from fluoroscopy. Esophageal endoscopy provides
a direct view of the esophageal mucosa and allows
biopsy of ulcerations and areas suspicious for malig-
nancy; however, histological evidence of esophagitis is
not sensitive for diagnosing reflux disease. Hydrochloric
acid (0.1 N) infused into the distal esophagus (Bernstein
acid infusion test) can reproduce symptoms of reflux
and confirm their esophageal origin.
The Tuttle acid reflux test is considered the reference
standard but is technically demanding. Reflux events are
detected by positioning a pH electrode in the distal
esophagus and monitoring over 24 hours. An abrupt
drop in pH (<4) is diagnostic of a reflux event. However,
a second reflux event cannot be detected until acid has
cleared the pH probe. Although reflux volume clears
within seconds, acid clearance takes several minutes
because neutralization by swallowed saliva is necessary.
This limits its temporal sensitivity for detection of reflux.
Scintigraphy
The radionuclide method is the most sensitive noninva-
sive method for detecting gastroesophageal reflux and
352 NUCLEAR MEDICINE: THE REQUISITES
Figure 11-3, cont'd B, Two minute frames × 22. Persistent retention in distal esophagus with
minimal evidence of clearance into the stomach.
has a low radiation exposure compared with barium
studies.It is physiological,easy to perform,well-tolerated,
and quantitative. Scintigraphic results are most affected
by the volume of the ingested meal and the rate of gastric
emptying. Sensitivity for detection of reflux decreases as
the stomach empties.
Methodology
In the past, the radionuclide reflux study was performed
differently in adults and children. The adult technique was
done in a manner similar to barium contrast methodology,
using Valsalva maneuvers and an abdominal binder to pro-
gressively increase intra-abdominal pressure. Static 30-sec-
ond images were acquired. Sensitivity for detection of
reflux was not particularly high and the methodology was
not physiologic. This method is no longer recommended.
Continued
The standard pediatric reflux study is acquired with
a rapid framing rate of 5–10 seconds during ingestion of the
infant’s usual formula or milk meal (“Milk study”). A typical
protocol is described in Box 11-4. The high temporal acqui-
sition rate increases sensitivity for detection of reflux events
compared to the older method of 30–60 second acquisi-
tions. The same methodology is recommended for adults
as well as children,although with orange juice.
Image Interpretation
All frames should be reviewed individually or in cine-
matic display. Gastroesophageal reflux is seen as distinct
spikes of activity into the esophagus ( Fig. 11-7). A sim-
ple semiquantitative method of interpretation is to grade
each reflux event as high-level or low-level (greater or
less than midesophagus), by its duration (e.g., lasting less
 Gastrointestinal System 353

or greater than 10 seconds),and by the temporal relation-

Figure 11-3, cont'd C, Quantification. 63% retention after 20
minutes (<5% is normal). After ingestion of water and carbonated
beverage, rapid emptying ensues.
ship to meal ingestion (events at low gastric volume
carry greater significance). The study is usually per-
formed for 60 minutes. Delayed imaging at 2 hours can
be used for calculation of gastric emptying. The total
number of reflux events can be summed in four cate-
gories: low-level, <10 seconds; low-level, ≥10 seconds;
high-level, <10 seconds; high-level, ≥10 seconds.
Alternatively, time–activity curves can be generated
and regions-of-interest drawn for the oropharynx, esoph-
agus, and stomach.Various quantitative indices have been
used ( Box 11-5). Peaks greater than 5% generally corre-
spond to reflux.
Normal values for neonates have never been estab-
lished. The greater the number of high events and the
longer the reflux events, the more severe is the disease.
Reflux events that occur with small gastric volumes have
more clinical significance because reflux is occurring
without the effect of the increased pressure of a full meal
volume and acid buffering.
Gastric emptying can be quantified by drawing
a stomach region-of-interest on computer for the initial
time 0,1-hour,and 2-hour images and is usually expressed
as the percent emptying. Between 40% and 50% empty-
ing of milk at 1 hour and 60–75% at 2 hours is generally
considered to be normal. The 2-hour emptying is consid-
ered more reliable than the 1-hour emptying. Pulmonary

Normal Achalasia Esophageal spasm

100%
Proximal
Percent peak count
Middle Distal

2 4 6 8 10 20 40 60 80 100 20 40 60 80 100
Time (sec)
Figure 11-4 Esophageal time–activity profiles: normal, achalasia, and esophageal spasm. Regions
of interest are drawn on computer for the proximal, middle, and distal esophagus and time activity
curves generated. Left: Normal subject. Bolus proceeds promptly sequentially from proximal to
distal esophagus. Middle: Achalasia. Retention predominantly in the lower esophagus. Right: Spasm,
uncoordinated contraction. Bolus shows poor progression through the esophagus.
354 NUCLEAR MEDICINE: THE REQUISITES

Pharynx
ROI

Cardia
Time

Figure 11-5 Generation of condensed dynamic images. In

each consecutive frame, the data in an esophageal region of
interest are compressed into a single column, displaying the
distribution of the tracer from the pharynx to the proximal
stomach for each 0.8-second interval. The columns are arranged
consecutively, generating a space and time matrix, with vertical
and horizontal dimensions representing spatial and temporal
activity changes.

aspiration should be looked for with the aid of computer

enhancement.
Salivagram is a more sensitive method of detecting
aspiration than the reflux study. It is essentially a varia-
tion of an esophageal transit study ( Box 11-2). A small
volume labeled bolus of radiotracer is placed in the
infant’s posterior pharynx and the study is acquired
with a rapid framing rate followed by static imaging
( Fig. 11-8).

Accuracy
The early reported poor sensitivity (60–70%) for
radionuclide esophageal reflux studies was the result
of the long framing rates used. More rapid framing
methods described previously have reported high
Figure 11-6 Condensed dynamic images. A, Normal
sensitivities of 75–100%. Scintigraphy is more sensitive esophageal swallow. Uninterrupted transit of the bolus down the
than barium studies and results in considerably esophagus. B, Achalasia. Persistent retention of activity in the
less radiation exposure to the child. The gold stan- esophagus with minimal clearance towards end of study.
dard is considered to be pH monitoring; however, its
limitations were discussed previously (e.g., 24-hour
monitoring and poor temporal resolution). The highest
sensitivity is achieved by using a combination of GASTRIC MOTILITY
scintigraphy and pH monitoring. However, scinti-
graphy is noninvasive and more commonly used The radionuclide gastric emptying study is the standard
clinically. method for evaluating gastric motility because the tech-
The reported sensitivity for detection of aspiration on nique is noninvasive, sensitive, accurate, quantitative, and
gastroesophageal reflux studies is quite low, from 0–25%. simple to perform in the nuclear medicine clinic.
However, the salivagram study can often demonstrate Nonradionuclide techniques have serious limitations.
aspiration when it is clinically suspected but the gas- Gastric intubation methods require serial aspiration.
troesophageal reflux study is negative. Marker-dilution techniques with duodenal recovery are
 Gastrointestinal System 355

Box 11-4 Gastroesophageal Reflux (Milk

Study): Protocol Summary

PATIENT PREPARATION
Overnight fast.

COMPUTER SETUP
Framing rate of 5 to 10 sec/frame for 60 min.

RADIONUCLIDE
Tc-99m sulfur colloid, 0.2 to 1 mCi (5 μCi/ml)

FEEDING MEAL
Infants: Normal feeding meal (formula or milk).
The radionuclide is mixed with half of the meal and
fed to the child. The second “cold”half of the meal is
then fed to the child. Orange juice for older children
and adults.

IMAGING PROCEDURE
After burping infant, place supine with gamma camera
positioned posteriorly and the chest and upper
abdomen in the field of view. Place radioactive marker
at the mouth for several frames. Figure 11-7 Gastroesophageal reflux. Sequential 5-second
frames show episode of long-lasting (greater than 15 seconds)
Acquire delayed image of chest. Review for aspiration
high-grade (higher than mid-esophagus) gastroesophageal reflux.
with computer enhancement.
Quantify reflux and gastric emptying

Box 11-5 Methods Used for Quantifying

cumbersome,disliked by patients,and the tubing may alter
the rate of emptying. Radiographic contrast studies detect
gross mechanical obstruction, but are insensitive to motil-
ity disorders of the stomach and are not quantitative.
Physiology
Anatomically, the stomach is a smooth-muscle hollow vis-
cus, anatomically divided into the cardia, fundus, body,
antrum, and pylorus ( Fig. 11-9). The mucosa contains
glands that produce hydrochloric acid and digestive
enzymes. Gastric motility is produced by an interaction
of muscular and neural activity with feedback regulation
from the small bowel. The proximal and distal portions
of the stomach have distinct functions.
The proximal stomach, or fundus, acts as a reservoir,
accepting large ingested solid and liquid volumes with
only minimal increase in pressure (receptive relaxation
and accommodation). The muscular contractions of the
fundus are not rhythmic but rather tonic in character,
producing a constant pressure gradient between the
stomach and duodenum, which moves the stomach con-
tents distally. Liquid emptying is the result of this fundal-
produced pressure gradient. Emptying is volume
dependent; the larger the volume, the more rapid the
emptying, thus the emptying pattern is exponential
Gastroesophageal Reflux in
Children
Mean value of the esophageal time–activity curve (TAC)
as a percentage of the initial gastric activity.
Reflux index, derived by integrating esophageal TAC
over 60 min and dividing by initial gastric activity.
Percent activity relative to gastric activity in a specified
episode multiplied by duration of the episode.
Number of episodes of high-level and low-level reflux
and their duration, <10 seconds or >10 seconds.
( Fig. 11-10). There is no delay before emptying begins.
Nutrients, salts, and acidity in the liquid all slow the rate
of emptying.
The distal stomach, or antrum, has a neural pace-
maker initiating rhythmic phasic contractions, which are
responsible for solid emptying. After ingestion of solid
food, muscular contractions sweep down the antrum in
a ringlike pattern, squeezing the food toward the
pylorus. Large food particles are not allowed to pass and
are retropelled back toward the antrum. The food parti-
cles become progressively ground up, eventually becom-
ing small enough (1–2 mm) to pass through the pyloric
sphincter. The pylorus,at the junction of the antrum and
356 NUCLEAR MEDICINE: THE REQUISITES

Gastroesophageal
junction
Fundus

Pyloric
Duodenum sphincter
Body

Antrum

Figure 11-9 Gastric anatomy and function. The proximal

stomach (fundus) accommodates and stores food and is
responsible for liquid emptying. The distal stomach (antrum) is
responsible for solid emptying. The antrum mixes and grinds food
into small enough particles that they can pass through the pylorus.

0
Percent emptying

50

100
0 10 20 30
Figure 11-8 Salivagram: aspiration. Neonate with neurological
problems and swallowing difficulties. Prior gastroesophageal Time (min)
reflux (milk) study (not shown) revealed numerous reflux events, Figure 11-10 Normal clear liquid emptying. Ingestion of
but no aspiration. A, After radiotracer was placed in the posterior 300 ml of water with Tc-99m sulfur colloid. One-minute frames for
pharynx, sequential 5-second frames showed transit into the 30 minutes. Time–activity curve generated by drawing a whole
tracheal bifurcation. B, High-count image at the end of 60 minutes. stomach region of interest on computer. Emptying began
Radiotracer remains at the tracheal bifurcation. No esophageal immediately. The clearance curve pattern is exponential with
transit to stomach. a half-emptying time of 9 minutes (normal <20 minutes).

duodenal bulb, acts as a sieve, regulating outflow of gas-
tric contents.
The time required for grinding food into small particles
before solid emptying begins is called the lag phase.
During this phase there is no emptying. The solid material
is converted into chyme through contact with acid and
peptic enzymes and mechanical grinding.Once solid emp-
tying begins,clearance into the duodenum occurs at a con-
stant rate, usually in a linear manner ( Fig. 11-11). The rate
of emptying depends on the size and contents of the meal.
 Gastrointestinal System 357

Fat, acid, protein, and carbohydrates all act to slow empty-

ing ( Box 11-6). An increasing volume of liquids slows the
rate of solid emptying; however, the liquids empty faster
than the solid meal component ( Fig.11-12).
In the fasting state between meals there are forceful
contractions that empty nondigestible debris from the
stomach. Motilin, a peptide hormone secreted by the
upper small bowel mucosa, initiates this interdigestive
event. Gastric motility is further modulated by sympa-
thetic and parasympathetic neural innervation and a vari-
ety of hormones with complex interactions.

Gastric Stasis Syndromes

The radionuclide gastric emptying study cannot differ-
entiate an anatomical obstruction from functional gas-
troparesis. Mechanical causes of gastric stasis, such as
obstruction by tumor or pyloric channel ulcer, have to
be excluded by endoscopy or contrast barium radiog-
raphy.
Acute gastroparesis is seen in a variety of clinical situa-
tions, including viral gastroenteritis, trauma, and meta-
bolic derangements ( Box 11-7). The causes for chronic
gastric stasis are diverse and have been classified into
various categories (e.g., electrical, metabolic, neurologic,
and systemic) ( Box 11-8). Various commonly used thera-
peutic drugs may cause delayed gastric emptying, such as
anticholinergics, antidepressants, oral contraceptives,
and beta adrenergic agonists ( Box 11-9). Furthermore,

Box 11-6 Factors that Affect the Rate

of Gastric Emptying
0
PHYSIOLOGICAL
Meal content
Fat, protein, acid, osmolality
Percent emptying

Volume
Weight
50 Caloric density
Particle size
Time of day
Patient position (standing, sitting, supine)
Gender
Metabolic state
Stress
100
0 10 30 60 Drugs
B Time (min)
TECHNICAL
Figure 11-11 Normal solid gastric emptying. A, Five-minute
sequential images acquired for 60 minutes. The egg-sandwich meal Radioisotope decay
moves from the gastric fundus to the antrum in a normal manner. Attenuation correction method
Radioactive marker placed at the right upper chest to check for Scatter and septal penetration
motion. B, Computer-generated time-activity curve with delay (lag Single-head versus dual-head camera
phase) of 9 minutes (arrow) before emptying begins (normal Frequent versus infrequent image acquisition
5–25 minutes). A linear pattern of emptying follows. Greater than Method of quantification
50% emptying occurred by 90 minutes (normal, >35%).
358 NUCLEAR MEDICINE: THE REQUISITES

0 0
Solid

Percent emptying
Percent emptying

Solid

Liquid

Liquid

100 100
0 90 0 90
A Time (min) C Time (min)
0 0 Solid

Percent emptying
Percent emptying

Solid Liquid

Liquid

100 100
0 90 0 90

B Time (min) D Time (min)

Figure 11-12 Diabetic gastroparesis: dual-phase solid-liquid emptying. A, Normal nondiabetic

subject: Linear solid emptying after a short lag phase and prompt exponential liquid emptying.
B, Diabetic with normal solid and liquid emptying. C, Diabetic with delayed solid but normal liquid
emptying. D, Diabetic with delay in solid and liquid emptying.

nicotine, opiates, and alcohol all inhibit gastric emptying. gastric emptying,rather than delayed emptying,occurs in
Rapid gastric emptying is seen in patients who have had some diabetics.
prior surgery (e.g., pyloroplasty and gastrectomy) or dis- Delayed gastric emptying may also be caused acutely
eases such as gastrinoma and hyperthyroidism (Box 11-10). by hyperglycemia per se. Gastric motility studies thus
Gastroparesis is clinically manifested by symptoms should ideally be performed with the patient under opti-
that include early postprandial satiety, bloating, nausea, mal diabetic control.
and vomiting. Mild-to-moderate disease may be asymp-
tomatic. As the disease progresses, symptoms become Nonulcer Dyspepsia
clinically manifest. Rapid gastric emptying can also pro- A common clinical syndrome evaluated by gastroenterol-
duce symptoms (at times severe) including palpitations, ogists is the entity of nonulcer dyspepsia. It is character-
diaphoresis, weakness, and diarrhea (dumping syn- ized by upper abdominal symptoms, ulcerlike in some
drome). patients but dyspeptic in others. To make the diagnosis
of nonulcer dyspepsia, organic conditions must have
Diabetic Gastroparesis been excluded by radiologic and endoscopic studies. It is
Long-standing insulin-dependent diabetes mellitus is one considered a functional disease and some of the patients
of the more common causes for chronic gastroparesis. have delayed gastric emptying.
Pathophysiologically, it is the result of vagal nerve dam-
age as part of a generalized autonomic neuropathy. In Pharmacological Therapy
addition to producing disagreeable postprandial symp- Various drug therapies are used to treat gastroparesis.
toms,abnormal emptying may exacerbate the problem of Their prokinetic properties are mediated by differ-
diabetic glucose control because timing of the insulin ent mechanisms. Metoclopramide ( Reglan) has both
dose with food ingestion and absorption is critical. Rapid central and peripheral antidopaminergic properties.
 Gastrointestinal System 359

Box 11-7 Causes of Functional Gastric
Stasis Syndromes—Acute and
Chronic
Acute Dysfunction
Trauma
Postoperative ileus
Gastroenteritis
Hyperalimentation
Metabolic disorders:
hyperglycemia, acidosis,
hypokalemia, hypercalcemia,
hepatic coma, myxedema
Physiological effects:
labyrinth stimulation,
physical and mental stress,
gastric distention, increased
intragastric pressure
Hormones: gastrin,
secretin, glucagon,
cholecystokinin,
somatostatin,
estrogen, progesterone
Box 11-8 Common Causes of Delayed
Gastric Emptying
NEUROMUSCULAR
Chronic Diseases ANATOMIC DISORDERS
Peptic ulceration Diabetic gastroparesis
Diabetes mellitus Surgery/Vagotomy Smooth muscle disorders
Hypothyroidism Pyloric hypertrophy Ileus
Progressive systemic Postradiotherapy Systemic diseases
sclerosis Tumors Scleroderma
Systemic lupus
erythematosus METABOLIC DISORDERS AMYLOIDOSIS
Dermatomyositis Electrolyte disturbances Anorexia nervosa
Familial dysautonomia Diabetic acidosis Acute viral infections
Pernicious anemia Uremia
Bulbar poliomyelitis
Amyloidosis
Gastric ulcer
Postvagotomy
Tumor-associated
gastroparesis Box 11-9 Drugs that Delay Gastric
Fabry disease
Emptying
Myotonic dystrophy
Drug Type Specific Drugs

Cardiovascular Nifedipine, beta adrenergic agonists

Respiratory Isoproterenol, theophylline
Gastrointestinal Sucralfate, anticholinergics
Neuropsychiatric Levodopa, diazepam, tricyclic
antidepressants, phenothiazine
Neurological side effects (e.g., drowsiness, lassitude) Reproductive Progesterone, oral contraceptives
are not uncommon. Cisapride (Propulsid) works by Drug abuse Alcohol, nicotine, opiates
releasing acetylcholine from the myenteric plexus.
Domperidone, not available in the United States, is
a peripheral dopamine antagonist that penetrates the
blood–brain barrier poorly, rarely producing neurological
side effects. Erythromycin is a motilin agonist with pro-
kinetic properties.Nausea as a side effect is common. All
improve emptying by increasing the amplitude of antral Box 11-10 Causes of Rapid Gastric
contractions. Anticholinergics can be effective in slow- Emptying
ing rapid emptying and ameliorating the symptoms (e.g.,
in patients with pylorospasm). POSTOPERATIVE
The side effects of therapeutic drugs discussed previ- Pyloroplasty
ously are reasons to confirm the diagnosis of abnormal Hemigastrectomy (Billroth I, II)
gastric motility before treating.
DISEASES
Duodenal ulcer
Scintigraphy Gastrinoma (Zollinger-Ellison syndrome)
The radionuclide gastric emptying study is the Hyperthyroidism
accepted standard for evaluation of gastric transit. HORMONES
Various other methods have been investigated, but
Thyroxine
none provide this physiological information so accu-
Motilin
rately, reproducibly, and are so relatively simple to Enterogastrone
perform.
360 NUCLEAR MEDICINE: THE REQUISITES
Clinical Indications
Recommendations have been published regarding gener-
ally agreed-upon indications for a radionuclide gastric
emptying study ( Box 11-11). They include diabetic
patients with symptoms of gastroparesis (particularly
those with poor glucose control, nonulcer dyspepsia,
severe reflux esophagitis,unexplained nausea,and vomit-
ing) and to assess response to a motility drug.
Radiopharmaceuticals
For accurate quantification of solid gastric emptying, the
radioactive tracer must be tightly bound to the ingested
food. Elution of the radiolabel will result in a part-solid,
part-liquid labeled mixture resulting in an erroneously
shortened solid emptying time because liquids empty
faster than solids.
A physiologically superb in vivo method of radiola-
beling chicken liver was described and used by early
investigators. Tc-99m SC was injected into the wing
vein of a chicken. After extraction by Kupffer cells of
the liver where it is fixed intracellularly, the chicken
was sacrificed and the liver was removed and cooked.
The liver was mixed with beef or chicken stew for
palatability and volume. This radiolabel was highly sta-
ble and did not dissociate after ingestion. Although of
proven utility, this method is not generally used for
obvious reasons.
Alternative in vitro methods of labeling liver have
proven acceptable. Good radiolabeling results by cook-
ing (fry, grill, microwave, etc.) Tc-99m SC in liver pâté,
injecting it into liver cubes, or even surface-labeling it;
the radionuclide is bound within the meat.
However, in current practice, most nuclear medicine
clinics radiolabel eggs with Tc-99m SC. Frying or scram-
bling the eggs with Tc-99m SC produces good binding to
the egg albumin. Egg substitutes can be used. Often
administered as an egg sandwich, this semisolid meal is
easy to prepare and palatable to most patients. Liquid
meal markers should equilibrate rapidly and be nonab-
Box 11-11 Clinical Indications for
Radionuclide Gastric
Emptying Study
Insulin dependent diabetics with persistent
postprandial symptoms
Diabetics with poor blood glucose control
Nonulcer dyspepsia
Severe reflux esophagitis.
Unexplained nausea and vomiting
Assess response to a motility drug
sorbable. Tc-99m SC or Tc-99m DTPA in water meet
these criteria.
Two-phase markers, one for the solid meal and another
for the liquid phase, have been used. In these dual-
isotope, solid-liquid studies, In-111 DTPA is often used as
the liquid marker (171 and 247 keV) in combination with
the Tc-99m SC (140 keV) solid meal ( see Fig. 11-12).
Methodology
The radionuclide gastric emptying study is straightfor-
ward. After the patient ingests a radiolabeled meal, inter-
mittent or dynamic imaging is performed for the
duration of the study. On computer, regions of interest
are drawn around the stomach and gastric emptying is
quantified.
Numerous protocols have been used for radionuclide
gastric emptying. Meal composition, patient positioning,
instrumentation, frequency of data acquisition, study
length, and quantitative methods have varied (Table 11-3).
Because many of these factors affect the rate of empty-
ing, no general normal values can be used. In each clinic,
the test must be standardized (i.e., performed the same
way for all studies, at the same time of day, with the
same meal, instrumentation, and computer processing).
Normal values must be determined for the protocol used
or one must closely follow a protocol in the medical liter-
ature and use its normal values. Some of the factors that
affect the rate of gastric emptying and normal values are
listed in Box 11-6.
Meal
The food content of the ingested meal is the major factor
affecting the rate of gastric emptying.Clear liquids empty
faster than liquids containing nutrients. Liquids empty
faster than semisolids, which empty faster than solids.
Large meals empty faster than small meals. Increases in
volume, weight, caloric density, and particle size all slow
the rate of gastric emptying. Normal emptying values are
meal-specific. Normal values apply to a meal of a specific
type with a specific volume and calorie content.
Routine Solid versus Liquid Meal
The solid or semisolid meal is more sensitive than liquid
emptying for detection of abnormal gastric emptying.
Liquid emptying is always normal when solid emptying is
normal.When solid emptying is delayed, liquid emptying
may be normal or delayed, depending on the severity of
the gastroparesis ( see Fig. 11-12). A liquid-only study
should be reserved for patients who cannot tolerate
solids ( Fig. 11-13).
Single versus Dual Isotope
A dual-isotope study allows for simultaneous evaluation of
liquid and solid gastric emptying. Although dual-phase
may be important for the investigation of gastric phy-
siology and pharmacology, it adds complexity, cost, and
increased radiation exposure to the patient.Furthermore,
 Gastrointestinal System 361

Table 11-3 Gastric Emptying Protocols

Single-headed camera (LAO method)* Dual-headed camera (geometric mean)†

Preparation Overnight fast Overnight fast

Meal Tc-99m sulfur colloid egg white sandwich, Tc-99m sulfur colloid egg white sandwich,
200 ml water 200 ml water
Dose Tc-99m sulfur colloid, 1 mCi (37 MBq) Tc-99m sulfur colloid, 1 mCi (37 MBq)
Window 15% 140 keV 15% 140 keV
Patient position Semi-upright (60˚) Supine
Projections LAO Anterior and posterior simultaneously
Framing rate 90 sec/frame for 90 min 60 sec/frame for 90 min
Decay correct Yes Yes
Attenuation correction LAO nonmathematical method Geometric mean
Computer processing ROI around summed gastric image ROI around summed gastric image
Data presentation TAC of counts versus time TAC of anterior, posterior, and geometric mean
counts versus time
Quantification Percent emptying at 90 min Percent emptying at 90 min
Abnormal Less than 35% Less than 34%

LAO, Left anterior oblique; ROI, region of interest;TAC, time–activity curve.

*From Ziessman HA, Fahey FH, Collen MD: Biphasic solid and liquid gastric emptying in normal controls and diabetics using continuous acquisition in the LAO view. Dig
Dis Sci 37:744–750, 1992.
†From Ziessman HA, Fahey FH,Atkins FB,Tall J: Standardization and quantification of radionuclide solid gastric emptying studies. J Nucl Med 45:760–764, 2004.

A
LIQUID GASTRIC EMPTYING
1200
1000
800
Counts

600
400
200
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
B min
Figure 11-13 Delayed clear liquid gastric emptying. A, Sequential 1-minute images with
delayed emptying of water from the stomach. B, A region-of-interest drawn around stomach and a
time–activity curve generated. T12⁄ of 38 minutes (normal 10–20).
362 NUCLEAR MEDICINE: THE REQUISITES
the dual-isotope study has no added clinical benefit over
a single-isotope solid emptying study.
Study Length
Protocol length varies from clinic to clinic, ranging from
60 minutes to 4 hours, although 90–120 minutes is the
most common. To some extent, this decision depends
on the meal ingested. Large, difficult-to-digest meals
(stew) empty slowly and may require 2.5–3 hours for
acquisition, whereas smaller, more easily digestible semi-
solid meals (eggs) require a shorter study length, about
1.5 hours.
Recent publications have recommended that gastric
emptying studies can be simplified in methodology
while maximizing sensitivity by acquiring images at time
0, 2 hours, and 4 hours, and utilizing a large published
database for normal values. Further confirmation is
needed.
Decay Correction
Correction for radioactive decay is mandatory for Tc-99m
labeled meals because of the short half-life of the
radionuclide (6 hours) and the relatively long duration of
the study.
Attenuation Correction
The ingested radiolabeled meal moves from the gastric
fundus to the gastric antrum, which is more anteriorly
located. The posterior-to-anterior movement of the radi-
olabeled gastric contents results in variable attenuation.
A radiolabeled meal is detected with increasing effi-
ciency as it moves towards the gamma camera because of
the decreasing amount of attenuating soft tissue between
the camera and stomach contents.
In a gastric emptying study acquired with a single-
headed gamma camera placed anteriorly, the detected
radioactive counts often rise as the meal moves from the
fundus to the antrum, although the amount of food in the
stomach is unchanged. This artifact of attenuation
adversely affects the accuracy of quantification and
results in an underestimation of emptying. The amount
of error varies from patient to patient, depending on size
and anatomy. This artifact is a particular problem in
obese patients, but often occurs in nonobese patients as
well. The average error resulting from attenuation is
10–15%, but can be as high as 30–50% in some individu-
als. The percent error is unpredictable prior to the
study; thus, routine correction for attenuation should be
routinely performed to ensure accurate gastric emptying
quantification.
Geometric Mean Method Conjugate views are
used to reduce the variability in sensitivity for
radiotracer detection as a function of location in the
body. The geometric mean (GM ) method for atten-
uation correction is the accepted gold standard.
Opposed images are routinely acquired in the anterior
and posterior projections. Ideally, both images are
obtained simultaneously with a dual-headed gamma
camera.With a single-headed camera, the two opposing
views can be obtained sequentially. The GM method is
a mathematical correction performed at each data point
(Fig. 11-14).
Geometric mean (GM) = √ (countsanterior × countsposterior)
Left Anterior Oblique Method An alternative to the
GM method for attenuation correction is the left anterior
oblique (LAO) method. This technique compensates for
attenuation using a single-headed gamma camera and
allows for frequent image acquisition. It requires no
mathematical correction.
The rationale is that in the LAO oblique projection,the
stomach contents move roughly parallel to the head of
the gamma camera, thus minimizing the effect of attenua-
tion ( Fig. 11-15). The LAO method results correlate well
with the GM method ( Fig.11-16). The GM method is still
the standard and in general superior to the LAO method;
however, the LAO method is adequate for most clinical
purposes and definitely superior to the anterior view-
only acquisition.
Frequency of Image Acquisition
Frequent image acquisition (e.g., a 1–5 minute framing
rate) permits dynamic temporal visualization of empty-
ing. With infrequent image acquisition (e.g., 30–60
minute framing rate) considerable information is lost.
Potential sources of error may not be appreciated with
infrequent imaging (e.g.,gastroesophageal reflux,motion,
overlap of stomach, and small bowel activity). Fewer data
points are available for determining the rate of emptying.
Thus, there is the potential for quantitative errors.
Despite these potential problems, for many clinical
0
Anterior
Percent emptying
50
Geometric mean
Posterior
100
0 30 60 90
Time (min)
Figure 11-14 Geometric mean (GM) attenuation correction.
The time–activity curves of both anterior and posterior
projections show the effect of varying attenuation. Anterior data
has a rising time–activity curve before it begins to empty. Posterior
data shows decreasing counts from time zero. The GM curve is a
normal two-phase solid emptying pattern.

Figure 11-15 Left anterior oblique method. If the camera is
placed in the LAO projection, the radiolabeled stomach contents
move roughly parallel to the head of the gamma camera,
compensating for the effect of varying attenuation. No
mathematical correction is needed.
purposes, infrequent image acquisition provides gener-
ally reliable results.
Scatter Correction
With dual-isotope meals (e.g., In-111 and Tc-99m), correc-
tion for downscatter ( In-111 into the Tc-99m window)
and correction for upscatter ( Tc-99m into the In-111 171
keV window) may be necessary. If and how much correc-
tion is needed can be determined from a simulated phan-
tom study. However, the error is inconsequential when
the dose ratio of Tc-99m to In-111 is at least 4:1 to 5:1.
Quantitative Analysis
The best method for processing a gastric emptying study
depends to some extent on how the study is acquired
(e.g.,frequency of image acquisition,static versus dynamic
acquisition, anterior and posterior views acquired sequen-
tially or simultaneously).Generally,a gastric region of inter-
est is drawn on the computer for individual or summed
images. After correction for decay and attenuation,
a time–activity curve is generated and a parameter of gas-
tric emptying calculated,usually a half-time of emptying or
a percent emptying at the end of the study.
The specific protocol used is to some extent deter-
mined by factors unique to each clinic. Table 11-3
describes two typical and validated protocols with nor-
mal values.
Liquid Emptying Clear liquids have no lag phase.
Emptying begins immediately after ingestion. The
clearance time–activity curve is monoexponential, with
a normal half-time of emptying of 10–20 minutes ( Figs.
11-10 and 11-13). Liquids with nutrients empty slower
than clear liquids. Liquids in a dual-phase meal also empty
slower than a liquid-only meal (Fig.11-12).
Solid Emptying After an initial delay before emp-
tying begins (lag phase), solid meals usually empty in
a linear manner. The length of the lag phase is affected
Gastrointestinal System 363
by many of the same factors that affect the rate of
emptying.
Various different methods for quantifying solid empty-
ing have been used. From a physiological standpoint, one
might wish to calculate the length of the lag phase and
then a rate of emptying.However,the clinical importance
of the lag phase is uncertain. A few investigators have
reported that a prolonged lag phase is the cause of
delayed emptying in certain diseases and others have
reported that the prokinetic effect of certain drugs
improves emptying by shortening the lag phase.
However, data are limited and conflicting. Differences in
methodologies likely account for the discrepant results.
Further investigations with frequent image acquisition
are required to answer this question.
The rate of emptying is best calculated after the lag
phase using a curve fitting method to determine either
a linear ( %/min) or exponential rate of emptying (t1/2).
A linear rate of emptying would be more physiologically
correct in most cases, although the difference between
a linear and exponential fit may not be great in many cases.
Thus, one of two methods of quantification has been
generally used for clinical purposes, either a half-time of
emptying beginning from time 0 (start of the study) or
a percent emptying (at the end of the study) calculated at
the end of the study. These approaches incorporate
both phases of solid emptying into the overall result.
Evaluation of Therapeutic Effectiveness
The radionuclide gastric emptying study is uniquely able
to evaluate the effectiveness of therapy. A patient with
gastroparesis should have a repeat study after initiation
of prokinetic therapy to judge whether in fact emptying
has improved ( Fig. 11-17). In some cases, prokinetic
drugs improve the patient’s symptoms without discern-
able improvement in emptying. This knowledge would
be particularly important in insulin-dependent diabetic
patients. The success of various surgical interventions
(e.g., gastroplasty, surgical relief of obstruction) can also
be effectively judged ( Fig. 11-18).
HELICOBACTER PYLORI INFECTION
A major medical advance was the discovery that ulcer
disease was caused by a bacteria, which led to antibiotic
cure of the disease rather than the prior chronic medical
treatment of symptoms and, in many cases, surgery.
Helicobacter pylori, a gram-negative bacterium, is the
causative organism which infects the gastric mucosa of
patients with duodenal ulcer disease, gastric ulcer dis-
ease, and antral gastritis. Bacterial eradication greatly
decreases ulcer recurrence rates, reverses histological
changes, and promotes healing of active ulcers.
364 NUCLEAR MEDICINE: THE REQUISITES

Figure 11-16 Attenuation correction using both GM and LAO methods. A, Anterior-only
acquisition shows moderate rise in counts over 15 minutes (top). GM correction (middle). Lag phase
shortened to 8 minutes and gastric emptying quantification is increased. LAO method (bottom).
Results very similar to GM method. B, Anterior-only acquisition shows very long lag phase of 50
minutes (top). GM correction shortens the lag phase considerably, although there is still some initial
rise suggesting incomplete correction (middle). Percent gastric emptying is much improved. LAO
method shows similar results (bottom).

an expired air-filled balloon can be mailed for breath

Urea Breath Test
In the presence of the bacterial enzyme urease, orally
administered urea is hydrolyzed to carbon dioxide (CO2)
and ammonia. If the urea carbon is labeled with either
the stable isotope carbon-13 or radioactive beta-emitter
C-14, it can be detected in the breath as labeled CO2.
H. pylori is the most common urease-containing gastric
pathogen; therefore, a positive urea breath test can be
equated with H. pylori infection. The radioactive
method is the most cost-effective.
The urea breath test is now widely available. This
test is simple to perform, noninvasive, accurate, and
inexpensive. An onsite analyzer is not needed because
gas analysis. The overall accuracy is high. However,
false-negative results may occur because of the recent
use of antibiotics or bismuth-containing medications.
False-positive results occur in patients with achlorhy-
dria, contamination with oral urease-containing bacte-
ria, and colonization with another Helicobacter, such
as H. felis.
The initial diagnosis is usually made by gastric biopsy.
The urea breath test is most commonly used to deter-
mine the effectiveness of therapy against H. pylori.
Serological tests cannot determine the effectiveness of
therapy because the antibody titer falls too slowly to be
diagnostically useful.

Figure 11-17 Monitoring response to therapy. A, Baseline
delayed solid and liquid gastric emptying. B, With metoclopramide
treatment, both solid and liquid emptying improved. C, Because
the patient had persistent symptoms, cisapride therapy was given.
Both liquid and solid emptying improved further.
GASTROINTESTINAL BLEEDING
Effective and prompt therapy for acute gastrointestinal
bleeding depends on accurate localization of the site of
hemorrhage. The history and clinical examination can
often distinguish upper from lower tract bleeding (e.g.,
melena versus bright red rectal bleeding). Upper-tract
hemorrhage can be confirmed with gastric intubation and
localized with a flexible fiberoptic endoscopy. Lower-tract
bleeding is more problematic. During active hemorrhage,
colonoscopy and barium radiography are of limited value.
Contrast Angiography
Angiography is diagnostic when positive, but it is inva-
sive and will demonstrate the bleeding site only if the
Gastrointestinal System 365
100
Percent emptying
50
0 30 60 90
A Time (min)
100
Percent emptying
50
0 30 60 90
B Time (min)
Figure 11-18 Pyloric obstruction. A, Referred for postprandial
symptoms. Solid gastric emptying shows no emptying. Pyloric
obstruction secondary to peptic ulcer disease was subsequently
diagnosed and partial gastrectomy performed. B, Postoperative
study shows a short lag phase and relatively rapid emptying.
contrast agent is injected at the time of active hemor-
rhage. Gastrointestinal bleeding is typically intermittent
and the clinical evaluation of whether the patient is
actively bleeding at any one time is unreliable. Clinical
signs of active hemorrhage often develop after bleeding
has ceased.
Because repeated angiographic studies are not prac-
tical, it is often the angiographer who requests that
a radionuclide gastrointestinal bleeding study be per-
formed prior to the contrast study. The radionuclide
study ensures that the patient is indeed actively bleeding
and helps localize the bleeding site to the likely vascular
origin of the bleed (e.g.,celiac,superior,or inferior mesen-
teric) so that the angiographer can inject contrast
promptly into the appropriate artery,limiting the duration
of the study and the amount of contrast media required.
Radionuclide Scintigraphy
Tc-99m Sulfur Colloid
In 1977 at the University of Pennsylvania,Alavi et al. first
described using Tc-99m SC for scintigraphic detection of
366 NUCLEAR MEDICINE: THE REQUISITES
Bleeding site
Activity
Background
B Time
Figure 11-19 Tc-99m sulfur colloid gastrointestinal bleeding
study. A, After injection,Tc-99m SC is cleared rapidly by the
reticuloendothelial system, with a serum half-life of 3 minutes.
Tc-99m SC will extravasate at the site of bleeding with each
recirculation. Because of rapid background clearance, a high-
contrast image results. B, Time–activity curves demonstrate
rapid clearance of background and increasing activity at the
bleeding site.
gastrointestinal bleeding. The rationale for this method
is that after intravenous injection,Tc-99m SC is rapidly
extracted from the serum ( half-life of 3 minutes) by the
reticuloendothelial cells of the liver, spleen, and bone
marrow. By 15 minutes after injection, it is cleared from
the vascular system. During active bleeding, the radio-
pharmaceutical extravasates at the bleeding site into the
bowel lumen, increasing with each recirculation of
blood. Continued extravasation with simultaneous back-
ground clearance results in a high target-to-background
ratio, permitting visualization of the intra-abdominal
active bleeding site ( Fig. 11-19).
Methodology
Ten mCi (370 MBq) of Tc-99m SC are administered.
Flow images (1–2 seconds/frame) are followed by
serial 500k–750k count images of the abdomen and
pelvis acquired every 1–2 minutes for 20–30 minutes
( Box 11-12).
Image Interpretation
Rapid bleeding may be detected on the initial blood flow
images.Vascular blushes of tumors, angiodysplasia, and
Box 11-12 Tc-99m SC Scintigraphy for
Gastrointestinal Bleeding:
Protocol Summary
PATIENT PREPARATION
None
RADIOPHARMACEUTICAL
Tc-99m sulfur colloid, 10 mCi (370 MBq)
INSTRUMENTATION
Camera: Large-field-of-view gamma camera. Collimator:
high resolution, low energy, parallel hole. Interface
with nuclear medicine computer; intensity set so that
bone marrow can be seen
Computer setup: 1-sec/frame anterior flow images
obtained for 1 min, then acquire 500k to 750k images
of the abdomen every 1 to 2 min for 20 min
PATIENT POSITION
Supine; entire abdomen and pelvis in field of view
IMAGING PROCEDURE
Inject radiopharmaceutical intravenously.
Acquire on computer.
Acquire anterior views. Oblique, lateral, and posterior
views may be obtained as needed to confirm the site
of bleeding.
If no bleeding site is detected, obtain a 1000k count
image of the upper abdomen with oblique views to
evaluate the hepatic and splenic flexures. If negative,
repeat views of the lower abdomen 15 min later to
check for activity that may have been obscured in the
hepatic and splenic flexures.
If the scan is negative and recurrent active bleeding is
suspected, a repeat dose of Tc-99m sulfur colloid is
given and the protocol repeated.
arteriovenous malformations may be seen in the absence
of active bleeding. Active hemorrhage is most often
detected in the first 5–10 minutes of imaging on the
static high-count images ( Fig. 11-20).
Active bleeding will move in an intestinal tract pat-
tern, usually antegrade, but retrograde movement is not
uncommon. A fixed region of radiotracer accumulation
is not diagnostic of bleeding and likely is due to fixed
uptake of the Tc-99m SC (e.g., ectopic spleen, renal trans-
plants, and bone marrow) rather than intraluminal hem-
orrhage. Asymmetrical bone marrow uptake may be due
to marrow replacement by tumor, infarction, or fibrosis,
making the adjacent marrow appear as focal accumula-
tion and perhaps suggesting a bleeding site.
Detection of bleeding in the region of the splenic flex-
ure or transverse colon is complicated by intense normal
liver and spleen uptake.Oblique views may help. A major

Figure 11-20 Tc-99m sulfur colloid study: descending colon
bleed. A, Two 3-second flow images; the second image indicates
extravasation at the site of bleeding ( large arrowhead).
B and C, Four sequential 5-minute images confirm the site of
bleeding. Images in C show movement to the more distal left colon
(small arrowheads).
disadvantage of the Tc-99m SC method is that bleeding
must be active at the time of injection, similar to angiogra-
phy. However, if the initial study is negative, a repeat injec-
tion can be administered and imaging repeated.
Tc-99m Red Blood Cells
In 1979 at Massachusetts General Hospital,Winzelberg
et al. described the Tc-99m-labeled red blood cells
( RBCs) method that is now in general use. The advan-
tage of this method over Tc-99 SC is that imaging can be
performed for a longer length of time, increasing the like-
lihood for detection of intermittent bleeding. The study
length is limited only by the physical half-life of Tc-99m
and the stability of the radiolabel.
A high-labeling efficiency ensures a high target-
to-background ratio and optimal image quality, which can
affect scan interpretation. Free (unbound) Tc-99m pertech-
netate is taken up by the salivary glands and gastric mucosa
and then secreted into the gastrointestinal tract,potentially
complicating interpretation.Various labeling techniques
(e.g., in vivo, modified in vivo [in-vivtro], and in vitro) with
different labeling efficiencies have been used (Box 11-13).
Gastrointestinal System 367
Box 11-13 Methods of Tc-99m Red Blood
Cell Labeling
IN VIVO METHOD (LABELING EFFICIENCY, 75–80%)
1. Inject stannous pyrophosphate.
2. Wait 10–20 min.
3. Inject Tc-99m sodium pertechnetate.
MODIFIED IN VIVO (IN VIVTRO) METHOD (LABELING
EFFICIENCY, 85–90%)
1. Inject stannous pyrophosphate.
2. Wait 10–20 min.
3. Withdraw 5 to 8 ml of blood into shielded syringe
with technetium-99m.
4. Gently mix syringe contents for 10 min at room
temperature.
IN VITRO (BROOKHAVEN) METHOD (LABELING
EFFICIENCY, 98%)
1. Add 4 ml of heparinized blood to reagent vial of
2 mg Sn+2, 3.67 mg Na citrate, 5.5 mg dextrose, and
0.11 mg NaCl.
2. Incubate at room temperature for 5 min.
3. Add 2 ml of 4.4% EDTA.
4. Centrifuge tube for 5 min at 1300 g.
5. Withdraw 1.25 ml of packed red blood cells and
transfer to sterile vial containing Tc-99m.
6. Incubate at room temperature for 10 min.
IN VITRO COMMERCIAL KIT (ULTRATAG) (LABELING
EFFICIENCY, >97%)
1. Add 1–3 ml of blood (heparin or acid citrate
dextrose as anticoagulant) to reagent vial
(50–100 μg stannous chloride, 3.67 mg Na citrate)
and mix. Allow 5 min to react.
2. Add syringe 1 contents (0.6 mg sodium hypochlorite)
and mix by inverting four or five times.
3. Add contents of syringe 2 (8.7 mg citric acid, 32.5
mg Na citrate, dextrose) and mix.
4. Add 370–3700 MBq (10–100 mCi) of Tc-99m to
reaction vial.
5. Mix and allow to react for 20 min, with occasional
mixing.
The in vitro method is preferable because of its superior
labeling efficiency (>97%).
Radiolabeling Red Blood Cells
Tc-99m RBC labeling requires that the technetium-99m
be reduced intracellularly so that it will bind with the
intracellular protein of the beta chain of the hemoglobin
molecule. The stannous ion in the form of stannous
pyrophosphate (tin) is used for this purpose. The dose is
optimized to maximize the percentage of technetium
bound inside the erythrocyte and minimize extravascular
and circulating free technetium.
368 NUCLEAR MEDICINE: THE REQUISITES
A simple kit technique for labeling red blood cells
in vitro is commercially available ( UltraTag, Mallinckrodt,
St. Louis, MO) ( Fig. 11-21). This method uses whole
blood and does not require centrifugation or transfer of
erythrocytes. The in vivo and modified in vivo methods
depend on biological clearance of undesirable extracellu-
lar reduced stannous ion. The original in vitro method
removed it by centrifugation. The in vitro commercial
kit method prevents extracellular reduction of stannous
ion by adding an oxidizing agent (sodium hypochlorite),
which cannot enter the red blood cells.
Drugs, intravenous solutions, and various clinical con-
ditions may interfere with red blood cell labeling.
Heparin is the most common cause. It oxidizes the stan-
nous ion and complexes with pertechnetate, thus reduc-
ing labeling efficiency. Direct injection of tin or pertech-
netate into intravenous lines containing heparin should
be avoided. Dextrose, mannitol, and sorbitol and the
presence of antibodies to erythrocytes reduce labeling
efficiency.
Time
(min)
0
1. Add 1 to 3 ml of patient’s blood,
using heparin or ACD as anticoagulant,
and mix. Allow to react for 5 minutes.
5
6 Heparin
2. Add contents of syringe 1 and mix.
3. Add contents of syringe 2 and mix.
4. Add 370 to 3700 MBq (10 to 100 mCi)
sodium pertechnetate Tc-99m (in volume
of up to 3 ml) to reaction vial.
5. Mix and allow to react for 20 minutes
with occasional mixing.
26
Figure 11-21 In vitro erythrocyte labeling with Tc-99m
(UltraTag RBC). Each kit consists of three nonradioactive
components: a 10-ml vial containing stannous chloride, syringe 1
containing sodium hypochlorite, and syringe 2 containing citric
acid, sodium citrate, and dextrose (ACD). Typical labeling
efficiency is >97%.
Various different methodologies have been used over
the years for labeling of red blood cells with Tc-99m ( Box
11-13). The in vivo method was the original methodol-
ogy.It was simple to perform but had suboptimal labeling
efficiency. A modified in vivo method was subsequently
developed and widely used because of its improved
labeling efficiency. However, today most nuclear medi-
cine clinics use an in vitro commercially available
method. It has high labeling efficiency and is simple to
prepare.
In Vivo Nonradioactive stannous pyrophosphate
(15 μg/kg body weight) is reconstituted with saline and
injected intravenously. After 15–30 minutes, Tc-99m
pertechnetate is injected intravenously. The per-
technetate diffuses across the erythrocyte membrane,
where it is reduced by the stannous ions administered
previously. The Tc-99m label binds to the beta chain of
hemoglobin.
Although the in vivo technique is simple, the labeling
yield is less than ideal, on the order of 80% but frequently
as low as 60–65%. Tc-99m activity not labeled to red
blood cells can contribute to background activity and also
reduces the number of counts available from the cardiac
blood pool. In some cases, the labeling fails dramatically
because of drug–drug interactions or other causes of poor
labeling ( Box 11-14). Special care is taken not to inject
through heparinized intravenous tubing. For these rea-
sons, most laboratories have adopted either the modified
in vivo approach or preferably the in vitro approach.
Excessive gastric,thyroid,and soft tissue background activ-
ity suggests poor labeling with free Tc-99m pertechnetate.
Box 11-14 Causes of Poor Tc-99m Red
Blood Cell Labeling
Drug Mechanism or Comment
Stannous ion Too little or too much will result in poor
labeling
A Tc-99m labeled heparin complex may
form when Tc-99m is injected through
a heparinized catheter
Methyldopa; Oxidation of stannous ion, diminution
hydralazine of reduction capacity
Doxorubicin Decreased labeling efficiency; effect
related to drug concentration
Quinidine May increase production of antibodies
to red blood cells
Iodinated Multiple mechanisms: lowering of
contrast stannous reduction capacity, altering
stannous distribution, competition for
erythrocyte binding sites between
Tc-99m and iodide media, alterations
in Tc-99m binding sites
 Gastrointestinal System 369

Modified In Vivo (In Vivtro) First, cold stannous

pyrophosphate is administered intravenously. After the Table 11-4 Dosimetry for Gastrointestinal Bleeding:
Tc-99m Red Blood Cells (RBCs) and
15- to 30-minute wait for equilibration of stannous ion
Tc-99m Sulfur Colloid
in RBCs, 3–5 ml of blood is withdrawn through an
intravenous line into a shielded syringe containing
Tc-99m pertechnetate and a small amount of either acid- Tc-99m RBCs Tc-99m Sulfur
(UltraTag) colloid
citrate-dextrose (ACD) solution or heparin. The blood is cGy/740 MBq cGy/370 MBq
incubated at room temperature for at least 10 minutes. (rads/20 mCi) (rads/10 mCi)
The syringe is agitated periodically and its contents are
Heart wall 2.0
reinjected into the patient. The syringe is left attached to
Bladder wall 0.48
the intravenous line during the procedure so that the Spleen 2.2 2.10
entire system is closed with respect to the patient’s Blood 0.80
circulation. The labeling efficiency in the modified Liver 0.58 3.40
in vivo approach is approximately 90%. Red marrow 0.30 0.27
Ovaries 0.32 0.06
In Vitro Blood is first withdrawn from the patient and
Testes 0.22 0.01
added to a reaction vial containing cold stannous chloride. Whole body 0.3 0.19
The stannous ion diffuses across the RBC membrane and
binds to the hemoglobin. Sodium hypochlorite is added to From package inserts: Mallinckrodt, St. Louis, MO.
oxidize excess extracellular stannous ion to prevent
reduction of Tc-99m pertechnetate. A sequestering agent
is added to remove extracellular stannous ion. Radioactive
labeling is then accomplished by adding Tc-99m
pertechnetate, which crosses the RBC membrane and is Box 11-15 Tc-99m Red Blood Cell
reduced by stannous ion in the cell. The mixture is Scintigraphy for Gastrointestinal
incubated for 20 minutes before reinjection. Labeling Bleeding: Protocol Summary
efficiency is greater than 95%. This approach is less
subject to drug–drug labeling interference and to problems PATIENT PREPARATION
of excessive or deficient stannous ion. A simple in vitro kit None
(UltraTag RBC,Mallinckrodt) is commercially available.
Dosimetry RADIOPHARMACEUTICAL
The radiation absorbed dose to the patient both using Tc-99m labeled red blood cells
the Tc-99m SC technique and the Tc-99m-labeled red
INSTRUMENTATION
blood cell method is relatively low, particularly when
compared with contrast angiography. The target organ Camera: Large-field-of-view gamma.
for Tc-99m SC is the liver; for Tc-99m erythrocytes, the Collimator: high resolution, parallel hole.
liver and the myocardial wall. The whole-body dose for Computer setup: 1-sec frames for 60 sec; 1-min frames
for 60 to 90 min.
labeled erythrocytes is 0.15 cG/925 MBq (0.4 rad/25
If needed: 2-4-hr delayed image sequence as 1-min
mCi) ( Table 11-4). frames for 20 to 30 min.
Methodology Static images: 2- to 3-sec flow images and 1000k count
A detailed imaging protocol is described in Box 11-15. images every 2 to 5 min.
An initial flow study is acquired at a 1–2 second framing Set intensity so that aorta, inferior vena cava, and iliac
rate ( Fig. 11-22). The patient is supine and the camera vessels are well visualized.
placed anteriorly. Then, 1-minute images are acquired
PATIENT POSITION
for 90 minutes ( Figs. 11-23 to 11-26). A left lateral view
of the pelvis can help differentiate activity in the bladder Supine; anterior imaging, with abdomen and pelvis in
from that in the rectum, and even occasionally, from field of view.
penile activity ( Fig. 11-27). If the study is not diagnostic, IMAGING PROCEDURE
delayed imaging can be obtained for up to 24 hours.
Inject patient’s Tc-99m-labeled erythrocytes
Delayed images should always be acquired at the same intravenously.
framing rate, usually for 30 minutes. Acquire flow images, followed by static images for
Image Interpretation 60 to 90 min.
Rapid hemorrhage may be seen on the flow phase of the If study is negative or bleeding is recurrent, repeat
study which can occasionally be helpful at a bleeding 30-min acquisition.
site difficult to see on later dynamic imaging ( e.g., adja-
370 NUCLEAR MEDICINE: THE REQUISITES
cent to the bladder). The flow phase can help localize
the site of bleeding even if bleeding is not active ( e.g.,
detection of a vascular blush due to angiodysplasia or
tumor) ( Fig. 11-22). Furthermore, vascular structures can
be defined ( e.g., kidneys, ectatic vessels, or uterus) to
help with image interpretation on delayed images.
Active bleeding is most often diagnosed by review of
the sequential 1-minute images obtained during the first
90 minutes of the study ( Figs. 11-23 to 11-26). Over 80%
of bleeding sites are detected during this initial imaging
time. Cinematic display on a computer can be very help-
ful for determining the pattern of flow (e.g., whether
small or large bowel). Because bleeding is intermittent,
delayed imaging beyond 90 minutes can sometimes be
helpful. The timing will depend on the patient’s condi-
tion and the logistics of the clinic.
Diagnostic Criteria
The purpose of the gastrointestinal bleeding study is to
diagnose active bleeding and to localize the site of bleed-
ing. Specific criteria must be used to avoid incorrect
interpretations ( Box 11-16). The radiotracer activity
must appear where there was none before, increase over
time, and conform to intestinal anatomy. The intralumi-
nal activity may move antegrade and/or retrograde.
Activity that is not moving should not be diagnosed
as an active bleeding site and is likely due to a fixed vas-
cular structure (e.g., hemangioma, accessory spleen, or
ectopic kidney).
Frequent image acquisition is important for localizing
the site of bleeding because hemorrhage can be rapid
and may move both antegrade and retrograde. Review of
1-minute dynamic frames displayed on a computer in cin-
ematic mode is very helpful for confirming and better
defining the site of bleeding.
Figure 11-22 Positive blood flow: no active bleeding. Increased focal blood flow to the region of
the ascending colon (arrowhead ). Ninety-minute Tc-99m red blood cell study was negative, as well
as a second acquisition for 30 minutes at 3 hours. Colonoscopy with biopsy diagnosed colon cancer
in the region of increased abnormal flow.
Carefully noting the pattern of intestinal transit of the
radioactivity is critical for determining the anatomical
bleeding site. An appreciation of gastrointestinal vascu-
lar anatomy and its embryological development is help-
ful in pinpointing the vascular bed for the angiographer
( Fig. 11-28).
Bleeding in the large intestines typically appears as
activity moving along the periphery of the abdomen in
elongated loops of bowel ( Figs. 11-23 and 11-24). The
small bowel is more centrally located and blood pro-
gresses rapidly through its curvilinear segments.
Localization of the site of bleeding to the small intestine
may be difficult ( Fig. 11-25). Slow small intestinal bleeds
may be hard to localize. They may first be detected by
pooling of radiotracer in the cecum. Intravenous
glucagon may be useful for inhibiting bowel peristalsis
and allow pooling of activity at the site of active bleeding
in the small bowel.
Because of scintigraphic limitations in exact anatomi-
cal localization of the bleeding site, it is often described
as in the general region of the small bowel, cecum,
ascending colon, hepatic flexure, transverse colon,
splenic flexure, descending colon, or rectosigmoid
colon. This is usually sufficient for the angiographer to
determine which vessel ( celiac, superior mesenteric, or
inferior mesenteric artery) to inject with contrast.
Although the radionuclide gastrointestinal bleeding
study is not generally ordered to diagnose upper gas-
trointestinal bleeding, the first indication of an upper
gastrointestinal source is sometimes revealed on the
bleeding scan ( Fig. 11-29). It is important to differenti-
ate upper gastrointestinal bleeding from free Tc-99m
pertechnetate. Free Tc-99m pertechnetate may simu-
late gastric bleeding due to uptake by the gastric

Figure 11-23 Tc-99m RBC: ascending colon bleed. Active hemorrhage originates in the
proximal ascending colon, then transits the transverse colon and enters the left colon by the end of
the 60-minute study. Etiology was colon cancer.
mucosa ( Fig. 11-29). Over time, activity may be seen to
move distal to the stomach in the small bowel and even
the large bowel on delayed images. Images of the thy-
roid and salivary glands should be obtained to confirm
or exclude free Tc-99m pertechnetate as a source of
gastric activity.
Pitfalls
Pitfalls are defined as normal, technical, or pathologi-
cal findings that may be misinterpreted active hemor-
rhage ( Box 11-17). The presence of free Tc-99m
pertechnetate caused by poor radiolabeling or dissoci-
ation of the label in vivo would be considered a techni-
cal pitfall. A common anatomical pitfall as a cause for
misinterpretation is focal activity in the genitourinary
tract. This could be due to pelvic or focal ureteral
retention either free pertechnetate or another Tc-99m
labeled reduced compound. This activity will usually
Gastrointestinal System 371
move with time, possibly making interpretation even
more problematic. Upright positioning, oblique, or
posterior views may be helpful to clarify the etiology.
Renal transplants and ectopic kidneys may also be
causes for concern.
Activity seen in the region of the pelvis can be misin-
terpreted. The differential diagnosis should include
bleeding in the rectum, bladder activity, uterine activity
during menses, and normal penile blood pool. A left lat-
eral view is essential for making the correct interpreta-
tion ( Figs. 11-26 and 11-29).
Intraluminal radioactivity first detected on delayed
images can pose a diagnostic dilemma. Blood first seen
in the sigmoid colon or rectum on a single delayed
image obtained between 8 to 24 hours may have origi-
nated from anywhere in the gastrointestinal tract.
Misinterpretation of this isolated finding can be avoided
372 NUCLEAR MEDICINE: THE REQUISITES
Figure 11-24 Tc-99m RBC: left colonic bleed. Dynamic images acquired over 60 minutes show
increasing activity in the region of the descending colon which moves distally to the sigmoid colon.
by acquiring dynamic 1-minute images whenever
delayed imaging is performed. An active bleeding site
should be diagnosed only by using the criteria already
described ( Box 11-16).
A pathological pitfall might include abdominal
varices, hemangioma, accessory spleen, arterial grafts,
and aneurysms (Fig. 11-30). Activity clearing through
the hepatobiliary system and gallbladder may be related
to hemobilia; however, patients with renal failure have
gallbladder visualization because of radiolabeled frag-
mented heme breakdown products (e.g., porphyrins).
Accuracy
Only 2–3 ml of extravasated blood is necessary for detec-
tion. In experimental studies, bleeding rates as low as
0.05–0.1 ml/min could be detected. This compares favor-
ably with the ability of contrast angiography to detect
bleeding rates of 1 ml/min or greater, at least a 10-fold dif-
ference.
Many investigations over the years have reported gen-
erally high accuracy for radionuclide gastrointestinal
bleeding scintigraphy; however, there are reports that
have found lower accuracy and have not found the study
helpful ( Table 11-5). Thus, controversy exists regarding
its clinical utility.
The reasons for the discrepancy in reported accuracy
of localization are several. Misinterpretation may be from
infrequent image acquisition, pitfalls described previ-
ously, and incorrect localization based on single delayed
images. However, other factors are likely. Patient referral
bias is one. Whether the patients had the radionuclide
bleeding study early in the course of their workup or
only after hospitalization with extensive negative radio-
logic and endoscopy evaluation is a critical factor in bias-
ing the investigation. The scintigraphic study will have
the highest yield when performed soon after arrival in
the emergency room or on admission.
The gold standard is often a problem with these inves-
tigations. The majority of patients do not get angiogra-
phy, and when they do, it may be a false-negative because
the patient is no longer actively bleeding. Colonoscopy is
not usually possible during active bleeding. Detection of
pathological abnormalities on radiographic studies or
colonoscopy after bleeding has ceased does not neces-
sarily indicate that they were the source of bleeding.
Overall, the majority of investigations have found the
gastrointestinal bleeding study is accurate in bleeding
site localization and therefore is clinically useful.
A telling point is that angiographers are the ones at many
institutions that aggressively demand scintigraphy prior
to their invasive procedure. Being available when needed
and having good communication with referring physi-
cians is critical for success.

Figure 11-25 Tc-99m RBC: duodenal bleed. Active bleeding is initially seen in the region of the
duodenum (arrowhead ) and sequential images show transit through the small intestines.
Tc-99m Red Blood Cells versus Tc-99m Sulfur
Colloid
The general consensus is that Tc-99m-labeled red blood
cell studies are superior to Tc-99m SC for diagnosing
acute gastrointestinal scintigraphy ( Table 11-6). A large
multicenter prospective study by Bunker et al. com-
pared these two approaches in 100 patients referred
with clinical evidence of acute bleeding. The Tc-99m SC
study was performed first, followed by labeled Tc-99m
red blood cells. Tc-99m SC localized only five sites of
hemorrhage, compared to 38 localized with Tc-99m RBC
imaging. The sensitivity and specificity of the Tc-99m
red blood cell study were 93% and 95%, respectively.
Continuous imaging for 90 minutes revealed 83% of all
active hemorrhages. Smaller studies have reported simi-
lar results.
The advantage of Tc-99m RBC scintigraphy is the abil-
ity to image over a longer time period. Tc-99m SC may
still have a limited role (e.g., in patients with active bleed-
ing who are clinically unstable). The quick preparation
Gastrointestinal System 373
time and 20-minute total imaging time may provide valu-
able information to the angiographer.
HETEROTOPIC GASTRIC MUCOSA
Meckel’s diverticulum is the most common and clinically
important form of heterotopic gastric mucosa. The ter-
minology often used is “ectopic” gastric mucosa;
however, it is not truly correct. Ectopic refers to an organ
that has migrated (e.g., ectopic kidney). Heterotopic
refers to a tissue at its site of origin. For example, other
manifestations of heterotopic gastric mucosa are in gas-
trointestinal duplications, postoperative retained gastric
antrums, and Barrett’s esophagus.
Radiopharmaceutical
Tc-99m pertechnetate has been used since 1970 to
diagnose heterotopic gastric mucosa. The most common
374 NUCLEAR MEDICINE: THE REQUISITES

Figure 11-26 Tc-99m RBC: rectal bleed. A, The last three of the 1-minute images show
increasing activity just superior and to the left of the bladder (arrowhead ) very suggestive of
rectosigmoid colon bleed. B, Left lateral view. Blood is seen in the region of the rectum (arrow).

Box 11-16 Criteria for Diagnosis of

Vascular Localization of Bleeding Site
supply: with Tc-99m Red Blood Cell
Celiac Scintigraphy
trunk

Focal activity appears

Superior Activity increases over time
mesenteric Activity movement conforms to intestinal anatomy
artery
Movement may be antegrade or retrograde
Inferior
mesenteric
artery clinical indication has been to detect a Meckel’s diver-
ticulum as the cause of gastrointestinal bleeding in
a child. However, it can be used to diagnose and local-
ize other types of heterotopic gastric mucosa listed
previously.

Figure 11-27 Vascular supply of gastrointestinal tract. The
embryological development of the gastrointestinal tract explains its
anatomical configuration and its vascular distribution. This schematic
diagram also relates the gastrointestinal anatomy to its arterial supply
(celiac,superior mesenteric,and inferior mesenteric arteries).
Mechanism of Uptake
The mucosa of the gastric fundus contains parietal cells,
which secrete hydrochloric acid and intrinsic factor, and
chief cells, which secrete pepsinogen. The antrum and
pylorus contain G cells, which secrete the hormone gas-
trin. Columnar mucin-secreting epithelial cells are found
throughout the stomach; they excrete an alkaline juice
that protects the mucosa from the highly acidic gastric
fluid.
Parietal cells were originally thought to be respon-
sible for Tc-99m pertechnetate gastric mucosal
uptake and secretion. Some experimental evidence sup-
ports this hypothesis; however, most evidence points to

Figure 11-28 Free Tc-99m pertechnetate. A gastrointestinal
bleeding study showed increasing gastric activity (not shown).
This subsequent image confirmed thyroid uptake, consistent with
free Tc-99m pertechnetate rather than active bleeding. The poor
target-to-background ratio is also due to considerable free
pertechnetate.
the mucin-secreting cells. Tc-99m pertechnetate up-
take has been found in gastric tissue with no parietal
cells and autoradiographic studies localize Tc-99m
pertechnetate uptake to the mucin cell rather than the
parietal cell.
A hypothesis explaining the conflicting data suggests
that the predominant mechanism is specific mucin cell
uptake and secretion,which is suppressible by potassium
or sodium perchlorate in a manner similar to iodide,
whereas parietal cell uptake is a minor factor, nonspe-
cific, secondary, and (as in chloride uptake) not sup-
pressed by perchlorate.
Dosimetry
The target organ for Tc-99m pertechnetate is the stom-
ach, followed by the thyroid gland ( Table 11-7).
Clinical Indications
Meckel’s Diverticulum
The most common congenital anomaly of the gastroin-
testinal tract is Meckel’s diverticulum, occurring in 1–3%
of the population. The diverticulum results from failure
Gastrointestinal System 375
of closure of the omphalomesenteric duct in the
embryo. The duct connects the yolk sac to the primi-
tive foregut through the umbilical cord. This true diver-
ticulum arises on the antimesenteric side of the small
bowel, usually 80–90 cm proximal to the ileocecal
valve. It is typically 2–3 cm in size but may be consider-
ably larger.
Heterotopic gastric mucosa is present in 10–30% of
patients with Meckel’s diverticulum, in approximately
60% of symptomatic patients, and in 98% of those that
manifest bleeding ( Box 11-18).
Clinical Manifestations
Gastric mucosal secretions can cause peptic ulceration
of the diverticulum or adjacent ileum, producing pain,
perforation,or most commonly,bleeding.Sixty percent of
patients with complications of Meckel’s diverticulum are
under age 2 years. Adults present with intussusceptions,
obstruction, infection, and abnormal fixation of the diver-
ticulum. Bleeding from Meckel’s diverticulum after age
40 is unusual.
Diagnosis
Meckel’s diverticulum is often missed on small bowel
radiography because it often has a narrow or stenotic
ostium, fills poorly, and has rapid emptying. Small bowel
enteroclysis is superior because the higher pressure of
the barium column more reliably fills the diverticulum.
Angiography is useful only with brisk active bleeding
and rarely used. The Tc-99m pertechnetate scan is con-
sidered the standard method for preoperative diagnosis
of a Meckel’s diverticulum.
Methodology
A typical protocol is described in Box 11-19. Patient
preparation is important. Because a full stomach or uri-
nary bladder may obscure an adjacent Meckel’s diverticu-
lum, fasting for 3–4 hours prior to the study or
continuous nasogastric aspiration to decrease the size of
the stomach is recommended and the patient should
void before, during, and after the study. Potassium per-
chlorate should not be used to block thyroid uptake
because it will also block uptake of Tc-99m pertechne-
tate by the gastric mucosa. It may be administered after
the study to wash out the radiotracer from the thyroid
and thus to minimize radiation exposure.
Barium studies should not be performed for several
days before scintigraphy because attenuation by the con-
trast material may prevent lesion detection. Procedures
(e.g., colonoscopy or laxatives that irritate the intestinal
mucosa) can result in nonspecific Tc-99m pertechnetate
uptake and should be avoided. Certain drugs (e.g., etho-
suximide [Zarontin]) may also cause unpredictable
uptake.
376 NUCLEAR MEDICINE: THE REQUISITES

Figure 11-29 Potential false-positive for gastrointestinal bleeding. A, Images acquired every 10
minutes × 6 show changing, increasing activity in the lower left and middle pelvic region. B, Anterior
(left) and left lateral (right) images acquired 90 minutes after tracer injection show the activity to be
the penile blood pool (arrowhead ). Left lateral views should be obtained whenever pelvic activity is
seen in order to separate rectal, bladder, and penile activity.

Pharmacological Augmentation
Various pharmacological maneuvers have been reported
to improve the detection of Meckel’s diverticulum,
including pentagastrin, glucagon, and cimetidine.
Cimetidine The histamine H2-receptor antagonist
cimetidine increases uptake of Tc-99m pertechnetate by
inhibiting its release from the gastric mucosa. No large or
controlled studies have been done to evaluate its
diagnostic utility; however, some recommend its routine
use because of its possible effectiveness and its lack of
significant risks or side effects. Others reserve its use for
a patient with a suspected false-negative Tc-99m
pertechnetate scan for Meckel’s diverticulum. The usual
dose is 20 mg/kg orally for 2 days prior to the study.
Pentagastrin and Glucagon Pretreatment with
pentagastrin has been used because it increases the
rapidity, duration, and intensity of Tc-99m pertechnetate
uptake. The mechanism may be the result of increased
acid production,leading to increased activity of the mucin-
producing cells and thus increased Tc-99m pertechnetate
 Gastrointestinal System 377

posterior location of renal or ureteral activity. Upright

BOX 11-17 Pitfalls in Interpretation of views may distinguish fixed activity (e.g., duodenum) from
Tc-99m Red Blood Cell ectopic gastric mucosa, which moves inferiorly, and this
Scintigraphy also serves to empty renal pelvic activity. The intensity of
activity may fluctuate because of intestinal secretions,hem-
PHYSIOLOGICAL orrhage, or increased motility washing out radiotracer.
Common Postvoid images can empty the renal collecting system and
Gastrointestinal (free Tc-99m pertechnetate)—stomach, aid in visualization of areas adjacent to the bladder.
small and large intestine Accuracy
Genitourinary False-negative studies can result from poor technique,
Pelvic kidney washout of the secreted Tc-99m pertechnetate, or lack of
Ectopic kidney sufficient gastric mucosa. Diverticula smaller than 2 cm2
Renal pelvic activity
may not be detectable by scintigraphy. An impaired
Ureter
blood supply to a diverticulum from intussusception,
Bladder
Uterine blush volvulus, or infarction can give false-negative studies.
Penis The reported accuracy of scintigraphy in the detec-
tion of Meckel’s diverticulum has varied, depending on
Uncommon the referral population studied (children or adults), the
Accessory spleen presenting symptom (rectal bleeding or abdominal
Hepatic hemangioma pain), and the technology used (old rectilinear scan-
Varices, esophageal and gastric ners or modern gamma cameras). However, one large
report summarized the results in 954 patients (mostly
RARE
children) who had undergone scintigraphy for sus-
Vascular pected Meckel’s diverticulum using modern imaging
Abdominal aortic aneurysm methods, and found an overall sensitivity of 85% and
Gastroduodenal artery aneurysm
a specificity of 95%.
Abdominal varices
Scintigraphy for Meckel’s diverticulum in adults
Caput medusae and dilated mesenteric veins
Gallbladder varices appears to have a poorer sensitivity than in children. One
Pseudoaneurysm series reported a sensitivity of only 63%. There were 10
Arterial grafts false-positive studies, although 7/10 of the subjects had
Cutaneous hemangioma surgically treatable disease. False negatives are probably
Duodenal telangiectasia related to the lack of gastric mucosa in the many adult
Angiodysplasia diverticula.
Causes for false-positive studies are listed in Box
MISCELLANEOUS
11-20. Normal structures may be confused with ectopic
Gallbladder (heme products in uremia) gastric mucosa if careful technique is not followed.
Factitious gastrointestinal bleeding
Activity in the genitourinary tract may cause false inter-
pretation. False positives may also result from inflamma-
tory or obstructive lesions of the intestines.
uptake. However, pentagastrin also increases intestinal
motility, leading to rapid movement into the small bowel. Gastrointestinal Duplications
Glucagon, because of its antiperistaltic effect, has been Duplications are cystic or tubular congenital abnormali-
used with pentagastric to decrease bowel peristalsis and ties that have a mucosa, smooth muscle, and an alimen-
prevent tracer washout from a diverticulum. However, tary epithelial lining attached to any part of the
pentagastrin is associated with significant side effects and gastrointestinal tract, but often the ileum. Most are symp-
is no longer commercially available in the United States. tomatic by age 2 years. Twenty percent are found in the
Image Interpretation mediastinum. Presenting symptoms are often those of
Scintigraphically, Meckel’s diverticulum appears as a focal Meckel’s diverticulum because 30–50% have heterotopic
area of increased intraperitoneal activity, most frequently gastric mucosa.
in the right lower quadrant (Fig. 11-31). Uptake is usually The diagnosis is usually made at surgery. Occasionally
first seen 5–10 minutes after tracer injection, increasing a preoperative diagnosis is made by barium radiography
over time at a rate similar to normal gastric uptake. or ultrasonography. Scintigraphy occasionally may be
Lateral or oblique views can be helpful in confirming helpful; for example, mediastinal gastrointestinal cysts
the anterior position of the diverticulum versus the have been diagnosed with Tc-99m pertechnetate.
378 NUCLEAR MEDICINE: THE REQUISITES

Figure 11-30 Aortic aneurysm and acute bleed. A, Flow study demonstrates a distal aortic
aneurysm. B, On dynamic imaging over 60 minutes,an acute bleed is seen to originate in the mid
lower pelvis,moving with time to the ascending and transverse colon,most consistent with a cecal
bleed. The midabdominal aortic aneurysm showed persistent activity from the beginning to the end
of the study. The fixed activity suggests that this is not active bleeding but anatomical.
 Gastrointestinal System 379

Table 11-5 Correct Localization of Gastrointestinal Table 11-7 Radiation Absorbed Dose for Tc-99m
Bleeding with Tc-99m Red Blood Cells Pertechnetate (Meckel’s Scan)

First author Year No. scans % Positive % Correct Target organ rad/mCi cGy/185 MBq (rad/5 mCi)

Suzman 1996 224 51 96 Bladder wall 0.053 0.265

Orechhia 1985 76 34 94 Stomach wall 0.250 1.250
O’Neill 2000 26 96 88 Large intestine wall 0.068 0.340
Emslie 1996 75 28 88 Ovaries 0.022 0.110
Leitman 1989 28 43 86 Red marrow 0.019 0.095
Bearn 1992 23 78 82 Testes 0.009 0.045
Dusold 1984 74 59 75 Thyroid 0.130 0.650
Rantis 1995 80 47 73 Total body 0.014 0.070
Van Geelen 1994 42 57 69
Nicholson 1989 43 72 67 From MIRD Primer, Society of Nuclear Medicine, Reston,VA, 1988.
Hunter 1990 203 26 58
Bentley 1991 182 60 52
Garofalo 1997 161 49 19
Voeller 1991 111 22 0
Box 11-18 Epidemiology of Meckel’s
Diverticulum

Table 11-6 Comparison of Tc-99m Sulfur Colloids 1–3% incidence in the general population.
(SC) and Tc-99m Red Blood Cells (RBCs) 50% occur by age 2 years.
for Gastrointestinal Bleeding 10–30% have ectopic gastric mucosa.
25–40% are symptomatic; 50–67% of these have ectopic
Tc-99m SC Tc-99m RBCs gastric mucosa.
95–98% of patients with bleeding have ectopic gastric
Dose 10 mCi (370 MBq) 25 mCi mucosa.
(may be repeated)
Dosimetry
Whole body 0.2 rad 0.4 rad
Target organ 3.6 rads (liver) 1.2 rads (heart)
Minimal bleeding 0.1 ml/min 0.05–0.4 ml/min through the gastrojejunostomy. The high acid produc-
detectable tion leads to marginal ulcers.
Labeling Commercial kit Commercial kit Endoscopy or barium radiography may demonstrate
Imaging duration 20–30 min 60–90 min the retained gastric antrum. Tc-99m pertechnetate
(repeat once) (repeat as needed
for 24 hr)
scintigraphy can be confirmatory. The protocol used is
Advantages Short imaging time Repeat imaging similar to that used for imaging a suspected Meckel’s
High target-to- up to 24 hr diverticulum. Uptake in the gastric remnant occurs simul-
background ratio taneously with gastric uptake and is seen as a collar of
Disadvantages Difficulty detecting False positive studies radioactivity in the duodenal stump of the afferent loop.
hepatic and splenic due to excretion
flexure bleeding of free Tc-99m
The retained antrum usually lies to the right of the gastric
Detects bleeding pertechnetate remnant. In one series,Tc-99m pertechnetate uptake was
only over short demonstrated in 16 of 22 patients with a retained antrum.
time
Barrett’s Esophagus
Chronic gastroesophageal reflux can cause the distal
esophagus to become lined by gastric columnar epithe-
Duplications often appear as large, sometimes multilobu- lium rather than the usual esophageal squamous epithe-
lated areas of increased activity. They are a cause for lium. This condition is known as Barrett’s esophagus
a false-positive interpretation of Meckel’s diverticulum. and is associated with complications of ulcers, strictures,
and an 8.5% incidence of esophageal adenocarcinoma.
Retained Gastric Antrum Tc-99m pertechnetate scans first demonstrated
The gastric antrum may be left behind in the afferent Barrett’s esophagus in 1973;however,today the diagnosis
loop after a Billroth II gastrojejunostomy. The antrum is typically made with endoscopy and mucosal biopsy.
continues to produce gastrin, which is no longer inhib- Positive scintigraphy shows intrathoracic uptake con-
ited by acid in the stomach because it has been diverted tiguous with the stomach but conforming to the shape
380 NUCLEAR MEDICINE: THE REQUISITES
Box 11-19 Meckel’s Diverticulum:
Protocol Summary
PATIENT PREPARATION
4–6 hr fasting before study to reduce size of stomach.
No pretreatment with sodium perchlorate; may be given
after completion of study.
No barium studies should be performed within 3 to 4
days of scintigraphy.
Void before, during if possible, and after study.
PREMEDICATION
Optional: Cimetidine, 20 mg/kg orally for 24 hours prior
to the study and 1 hour prior to the study.
RADIOPHARMACEUTICAL
Tc-99m pertechnetate
Children:30–100 μCi/kg (minimal dose 200 uCi) (7.4 MBq)
Adults: 5–10 mCi intravenously
INSTRUMENTATION
Camera: large-field-of-view gamma.
Collimator: low energy, all purpose or high resolution.
PATIENT POSITION
Supine under camera with xiphoid to symphysis pubis
in field of view
IMAGING PROCEDURE
Obtain flow images: 60 1-sec frames.
Obtain static images: 500k counts for first image, others
for same time every 5–10 min for 1 hr.
Erect, right lateral, posterior, or oblique views may be
helpful at 30–60 min.
Obtain postvoid image.
and posterior location of the esophagus. Scintigraphy
should be performed with the patient erect to minimize
reflux. To avoid misinterpreting accumulation in a hiatal
hernia as Barrett’s, the scan should be interpreted in con-
junction with an upper gastrointestinal contrast series.
False-negative results have been reported.
INTESTINAL FUNCTION AND TRANSIT
Schilling Test
Although most often ordered to diagnose pernicious ane-
mia, the Schilling test is really a test of vitamin B12
(methylcobalamin) absorption. Vitamin B12 can be
absorbed from the ileum only if it is complexed with
intrinsic factor, which is produced by gastric parietal
cells in the stomach. After absorption, vitamin B12 is
bound to storage sites in various tissues and slowly
metabolized.
Vitamin B12 deficiency manifests clinically as a mega-
loblastic anemia and neurological disease. The most com-
mon cause is an intrinsic factor (IF) deficiency in patients
with pernicious anemia and associated gastric atrophy.
Intestinal causes of vitamin B12 malabsorption include
Crohn’s disease, ileal resection, gluten enteropathy, tropi-
cal sprue, bacterial overgrowth syndromes, and fish tape-
worm (Diphyllobothrium latum) infestation. Pancreatic
insufficiency can also cause vitamin B12 malabsorption.
Vitamin B12 is usually labeled with cobalt-57 (Co-57 ).
Co-57 has a physical half-life of 272 days and a 122-keV
photopeak. An intramuscular flushing dose of unlabeled
vitamin B12 is administered to saturate tissue and plasma
binding sites,maximizing the renal excretion of absorbed
cobalt-labeled vitamin B12. Thus, the healthy person will
absorb the labeled vitamin and excrete it in the urine
through glomerular filtration. A urine collection is made
for 24 hours and sometimes 48 hours.
Renal excretion of cobalt is not continuous and one or
two missed samples in the collection could erroneously
result in an abnormal result. The urine volume should
be examined to ensure adequate function and collection.
The 48-hour sample may be used in cases of poor renal
function. A small sample is taken from the total collec-
tion and counted in a scintillation counter. The Schilling
test measures the fraction of the administered dose that
is excreted in the urine (normal, >9% in 24 hours).
The traditional approach first measures Co-57–labeled
vitamin B12 excretion (stage I). If it is abnormal, the study
is repeated (stage II) with the addition of IF. If excretion is
abnormal without IF but increases significantly with IF,the
diagnosis of pernicious anemia is confirmed. If both are
abnormal, pernicious anemia is ruled out and the cause is
small bowel malabsorption or pancreatic insufficiency.
An alternative second or third stage can be performed
(e.g., after antibiotic therapy for assumed bacterial over-
growth or with pancreatic enzyme replacement).
The second approach to the Schilling test administers
vitamin B12 and IF simultaneously, with Co-58–labeled
vitamin B12 and Co-57–labeled vitamin B12 bound to IF.
The advantage of this method is convenience. The disad-
vantage of the dual method is scatter,making calculations
more complex.
Intestinal Transit Scintigraphy
Small and large intestinal transit scintigraphy is not routine
in most laboratories and optimal methods are still under
investigation. This section is meant to provide a limited
overview of the subject. A detailed review of methodolo-
gies for intestinal transit studies is listed under Suggested
Reading (Maurer and Kevsky). Because a clinician may
inquire regarding such a study, clinical indications,
 Gastrointestinal System 381

Figure 11-31 Meckel’s diverticulum. A 7-year-old boy with rectal bleeding. Sequential images
show focal uptake in left lower quadrant confirmed at surgery to be a Meckel’s diverticulum
(arrowheads). Note simultaneous rate of uptake of the Meckel’s diverticulum and stomach.
Motion artifact can be seen in second image.

nonscintigraphic methodologies, and scintigraphic
methodologies are briefly reviewed.
Technical problems exist for accurately measuring
intestinal transit. The radiolabeled meal must withstand
the acidic environment of the stomach and the alkaline
milieu of the small bowel. Quantification is more com-
plex than for gastric emptying, where all the radiolabeled
meal resides in the stomach at the start of the study and
quantification depends only on the clearance rate.With
intestinal transit, there is a protracted infusion from the
stomach and no single time zero.
Small Bowel Transit
Chronic intestinal pseudo-obstruction, a disease of the
enteric nervous system, has delayed gastric and small
bowel motility. Small bowel transit may sometimes be
altered in irritable bowel syndrome, although primarily
a large bowel disease. Transit is typically faster through
the small and large bowel in patients with irritable bowel
syndrome with diarrhea and slower in patients with con-
stipation.
Nonscintigraphic Methods
The hydrogen breath test measures hydrogen produced
when a carbohydrate (C-14 lactose) is fermented by
colonic bacteria. The test measures the transit time of
the meal’s leading edge from mouth to cecum, but is not
an index of the transit of the meal’s bulk. The transit rate
is affected by the rate of gastric emptying. Fermentative
bacteria in the colon are required but absent in one
fourth of the population. The test is not widely available.
A small bowel barium follow-through study is not quan-
titative, although mixing barium with food and plotting
382 NUCLEAR MEDICINE: THE REQUISITES

orally, usually In-111 DTPA, with or without a solid meal.

Box 11-20 Causes for False-Positive Imaging for up to 4 hours is required. The radiolabel
Meckel’s Scan spreads out as it moves through the small bowel. That,as
well as anatomical variation (e.g., identification of the
URINARY TRACT cecum) make quantification challenging.Normal variabil-
Ectopic kidney ity in small bowel transit variation makes it difficult to
Extrarenal pelvis define normal values.
Hydronephrosis
Vesicoureteral reflux Large Bowel Transit
Horseshoe kidney Symptoms of colonic dysmotility include constipation,
Bladder diverticulum diarrhea, fecal incontinence and lower abdominal pain.
VASCULAR
Irritable bowel syndrome is the most common diagnosis
in patients with mild constipation.
Arteriovenous malformation
Several patterns of abnormal motility have been
Hemangioma
Aneurysm of intraabdominal vessel
described, including isolated anal sphincter dysfunction
Angiodysplasia (adult onset Hirschsprung’s disease), isolated rectosig-
moid dysfunction, slow transit through the entire colon
OTHER AREAS OF ECTOPIC GASTRIC MUCOSA (colonic inertia), and a generalized disorder of gastroin-
Gastrogenic cyst testinal dysfunction as seen in chronic intestinal pseudo-
Enteric duplication obstruction.
Duplication cysts Differentiation of these patterns can be clinically use-
Barrett’s esophagus ful. For example, prokinetic drugs can improve colonic
Retained gastric antrum inertia. If a surgical approach is indicated, the procedure
Pancreas will depend on whether the entire colon, rectosigmoid,
Duodenum
or only anorectal dysfunction is the problem.
Colon
Radiographic Methods
HYPEREMIA AND INFLAMMATORY Cineradiography and fluoroscopy with radiopaque mark-
Peptic ulcer ers have been used to estimate transit times. Interpre-
Crohn’s disease tation is limited by the infrequent abdominal images and
Ulcerative colitis difficulty in determining the exact location of the mark-
Abscess ers due to overlap of large and small bowel. The study is
Appendicitis not physiologic and results in a relatively large radiation
Colitis dose to the patient.
NEOPLASM
Large Bowel Transit Scintigraphy
The various different patterns of colonic transit described
Carcinoma of sigmoid colon
previously have been differentiated with scintigraphy.
Carcinoid
Lymphoma
The study requires 72–96 hours, giving the radiotracer
Leiomyosarcoma orally. In-111 DTPA has been most commonly used.
Gallium-67 citrate has also been used as a measure of
SMALL BOWEL OBSTRUCTION small and large intestinal transit. Given orally, Ga-67 is not
Intussusception absorbed from the bowel and 98% of the ingested dose is
Volvulus excreted in the feces.
Various other creative radiopharmaceuticals have
been investigated that will not breakdown before reach-
ing the large bowel (e.g., radiolabeled cellulose fiber). In-
its movement provides an index of the transit rate. The 111 polystyrene cation exchange resin pellets have been
radiation dosimetry is relatively high compared to scinti- placed in a gelatin capsule coated with a pH-sensitive
graphic methods and the meal is nonphysiological. polymer that resists disruption at pH levels of the stom-
Radionuclide Small Intestinal Transit ach and proximal small bowel but breaks down at the
Small bowel transit can be measured most rapidly and ileocecal valve because of the increasing pH.
accurately by direct placement of the radiotracer via intu- Various quantitative methods have been used (e.g.,
bation at the site of the proximal small bowel. However, orocecal transit times or initial arrival times). Cecal
this is invasive and not usually practical. Alternatively, arrival time has been defined as the time for accumula-
transit can be measured using a liquid marker taken tion of 10% of total abdominal counts in the cecum and

Figure 11-32 Protein-losing enteropathy. Patient received Tc-99m human serum albumin.
Immediate (A), 1-hour (B), and 2-hour (C) images. Increasing activity is seen in the small bowel
initially (B), with subsequent transit to the colon (C), consistent with protein-losing enteropathy.
ascending colon. More complicated methods of quantifi-
cation have generally been used (e.g., determining the
geometric center, which is a weighted average of the
counts in each region of the bowel).
Protein-Losing Enteropathy
Excessive protein loss through the gastrointestinal tract
has been associated with a variety of diseases, including
intestinal lymphangiectasia, Crohn disease, Menetrier dis-
ease, amyloidosis, and intestinal fistula. The resulting
hypoproteinemia can be a serious clinical problem.
Tc-99m human serum albumin, In-111 transferrin, and
Tc-99m dextran have all been successfully used to scinti-
graphically confirm a protein-losing enteropathy. In-111
chloride binds in vivo to serum proteins, most notably
transferring similarly to Ga-67, and abdominal imaging
can be used to visualize the protein leak. Serial abdomi-
nal images show radiotracer collection in the small
bowel in the first 30 minutes within increasing amounts
over 24 hours ( Fig. 11-32). None of these radiopharma-
ceuticals are available commercially in the United States.
SUGGESTED READING
Bunker SR, Lull RJ,Tanasescu DE, et al: Scintigraphy of gastroin-
testinal hemorrhage: superiority of 99mTc red blood cells over
99m Tc sulfur colloid. AJR 143:543-548, 1984.
Camilleri M, Hasler W, Parkman HP, et al: Measurement of gas-
trointestinal motility in the GI laboratory. Gastroenterology
115:747-762, 1998.
Castronovo FP, Jr: Gastroesophageal scintiscanning in a pedi-
atric population: dosimetry. J Nucl Med 27:1212-1214, 1986.
Datz FL: Considerations for accurately measuring gastric empty-
ing. J Nucl Med 32:881-884, 1991.
Diamond RH, Rothstein RD, Alavi A: The role of cimetidine-
enhanced Tc-99m pertechnetate imaging for visualizing
Meckel’s diverticulum. J Nucl Med 32:1422-1424, 1991.
Gastrointestinal System 383
Emslie JT, Zarnegar K, Siegel ME, et al:Technetium-99m-labeled
red blood cell scans in the investigation of gastrointestinal
bleeding. Dis Colon Rectum 39:750-754, 1996.
Fahey FH, Ziessman HA, Collin MJ, et al: Left anterior oblique
projection and peak-to-scatter ratio for attenuation compen-
sation of gastric emptying studies. J Nucl Med 30:233-239,
1989.
Heyman S: Pediatric nuclear gastroenterology: evaluation of gas-
troesophageal reflux and gastrointestinal bleeding. In Freeman
LM,Weissman HS (eds): Nuclear Medicine Annual. New York,
Raven Press, 1985.
Klein HA,Wald A: Esophageal transit scintigraphy. In Freeman
LM,Weissman HS (eds): Nuclear Medicine Annual. New York,
Raven Press, 1985.
Marianai G, Boni G, Barreca M, et al: Radionuclide gastro-
esophageal motor studies. J Nucl Med 45:1004-1028, 2004.
Maurer AH, Kevsky B:Whole-gut transit scintigraphy in the valu-
ation of small bowel and colon transit disorders. Sem Nucl Med
25:326-338, 1995.
Sfakianakis GN, Haase GM:Abdominal scintigraphy for ectopic
gastric mucosa: a retrospective analysis of 143 studies. AJR Am
J Roentgenol 138:7-12, 1982.
Tougas G, Eaker EY,Abell TL, et al:Assessment of gastric empty-
ing using a low fat meal: establishment of international control
values. Am J Gastroenterology 95:1456-1462, 2000.
Winzelberg GG: Radionuclide evaluation of gastrointestinal
bleeding. In Freeman LM (ed): Freeman and Johnson’s Clinical
Radionuclide Imaging, vol. 3. New York, Grune & Stratton,
1986.
Ziessman HA: Keep it simple—it’s only gastric emptying. In
Freeman LM (ed): Nuclear Medicine Annual. Philadelphia,
Lippincott Williams & Wilkins, 2000.
Ziessman HA, Fahey FH,Atkins FB,Tall J: Standardization and
quantification of radionuclide solid gastric-emptying studies.
J Nucl Med 45:760-764, 2004.
Zuckier LS: Acute gastrointestinal bleeding. Sem Nuclear
Medicine 33:297-311, 2003.
 12
CHAPTER Infection and
Inflammation

Pathophysiology of Inflammation and Infection Intra-abdominal Infection

Radiopharmaceuticals Inflammatory Bowel Disease
Gallium-67 Citrate Renal Disease
Chemistry and Physics Cardiovascular Disease
Mechanism of Uptake Pulmonary Infections
Pharmacokinetics and Distribution Sarcoidosis
Imaging Characteristics Idiopathic Interstitial Pulmonary Fibrosis
Methodology Pulmonary Drug Reactions
Normal Scintigraphy Immunosuppressed Patients
Dosimetry Diffuse Parenchymal Uptake
Radiolabeled Leukocytes Cytomegalovirus Infection
Leukocyte Physiology Focal Pulmonary Uptake
Indium-111 Oxine Leukocytes Negative Ga-67 Uptake
Technetium-99m HMPAO Labeled Leukocytes Intracerebral Infection
Technetium-99m Fanolesomab (NeutroSpec) Abdominal and Pelvic Infections
Technetium-99m Sulesomab (LeukoScan) Malignant External Otitis
Fluorine-18 Fluorodeoxyglucose Fever of Unknown Origin
Investigational Radiopharmaceuticals
Clinical Applications for Infection Scintigraphy Early identification and localization of infection is criti-
General Considerations cal for the appropriate and timely selection of therapy.
Osteomyelitis Computed tomography (CT ), magnetic resonance
Acute Hematogenous Osteomyelitis imaging (MRI), and ultrasonography are often first-line
Extension from a Contiguous Site of Infection imaging methods for detection and localization of infec-
Direct Introduction of Organisms into Bone tion. These methods are not always successful in find-
Clinical Diagnosis ing the site of infection, particularly if the patient has
Conventional Imaging no localizing symptoms or signs and if there is no focal
Scintigraphy fluid collection or morphological change. Postsurgical
Diabetic Foot or posttherapeutic anatomic changes can make detec-
Vertebral Osteomyelitis tion problematic.
Bone Scan Although developed in the early 1970s as a tumor-
Leukocyte Scintigraphy imaging agent, gallium-67 citrate (Ga-67) was noted to
Ga-67 Scintigraphy have an infection-seeking property. It remained the only
Other Scintigraphy infection-imaging radiopharmaceutical available until
Infected Joint Prostheses leukocytes labeled with indium-111 ( In-111) were
Bone Scan approved for clinical use in the mid-1980s. The next
Ga-67 Scintigraphy advance was radiolabeling leukocytes with technetium-
Leukocyte Scintigraphy 99m ( Tc-99m), with its better image quality and lower
Response to Therapy patient radiation.

384
 Infection and Inflammation 385

Today, both In-111- and Tc-99-radiolabeled leuko-

cytes are successfully used for infection scintigraphy. Box 12-1 Radiopharmaceuticals for
However, in vitro radiolabeling of leukocytes has draw- Infection Imaging
backs. Because of the time required for radiolabeling,
reinfusion of the cells typically cannot be accomplished CLINICAL USE
until 3–4 hours after their initial withdrawal. There is Gallium-67 citrate
also concern for the potential transmission of bloodborne Indium-111 oxine-labeled leukocytes
infections. Recently, a monoclonal antibody against sur- Technetium-99m HMPAO-labeled leukocytes
face antigens on granulocytes, Tc-99m fanolesomab Tc-99m fanolesomab (NeutroSpec)
( NeutroSpec), has been approved for clinical use. F-18 Tc-99m sulesomab (LeuTech) (approved in Europe)
fluorodeoxyglucose ( FDG) also has a potential role in F-18 fluorodeoxyglucose
infection imaging. New radiopharmaceuticals with vari- INVESTIGATIONAL
ous other mechanisms of uptake ( e.g., peptides,
Nonspecific IgG immunoglobulins
cytokines) are under investigation (Box 12-1, Table 12-1).
Tc-99m ciprofloxacin (Infecton)
Chemotactic peptides (interleukin-8)
Liposomes
PATHOPHYSIOLOGY OF Anti-E-selectin
INFLAMMATION AND INFECTION Lymphocytes

Inflammation is a response of tissues to injury that attracts

cells of the immune system, specialized serum proteins,

Table 12-1 Inflammation Terminology

Mast cell Tissue cell with basophilic granules containing vasoactive amines and heparin. Degranulates in response
to injury
Prostaglandins, leukotrienes Family of unsaturated fatty acids, components of most cell membranes. Responsible for induction of
pain, fever, vascular permeability, chemotaxis of neutrophils
Vasoamines Vasoactive amines that cause increased capillary permeability (e.g., histamine, 5-hydroxy-tryptamine,
produced by mast cells, basophils, and platelets)
Kinin system Series of serum peptides sequentially activated to cause vasodilation and increase permeability
Complement Cascading sequence of serum proteins, activated directly or via antigen–antibody interaction
C3a and C5a Stimulate release by mast cells of their vasoactive amines known as anaphylotoxins
Opsonization C3b attached to particle promotes sticking to phagocytic cells because of their C3 receptors. Antibody,
if present, augments this by binding to Fc receptors.
CRP Acute phase protein made in liver, appears in serum within hours of tissue damage or infection. Binds
to phosphorylcholine on bacterial surface, fixes complement, promotes phagocytosis.
Polymorphonuclear leukocyte The major mobile phagocytic cell, whose prompt arrival in tissues plays a vital part in removing
( PMN) invading bacteria
Monocyte Precursor of tissue macrophages ( MAC) responsible for removing damaged tissue and microorganisms.
Tissue macrophages are an important source of inflammatory cytokines.
Lysosomal enzymes Bactericidal enzymes released from the lysosomes of neutrophils, monocytes, macrophages
Inflammatory cytokines Inflammatory response is coordinated by several cytokines produced by various cell types, including TNF-α,
IL-6, and IL-1. All have many functions (e.g., initiating changes in vascular endothelium which promote
leukocyte entry into an inflammatory site). They induce acute phase response and tissue repair.
Chemotaxis C5a, C3a, leukotrienes and chemokines stimulate PMNs and monocytes to move into tissues.
Movement towards the site of inflammation is called chemotaxis
Chemokines Polypeptides that play a role in chemotaxis and regulation of leukocyte trafficking, e.g., interleukins (e.g., IL8).
Adhesion and cell traffic Change in expression of endothelial surface molecules, induced by cytokines, cause PMNs, monocytes,
and lymphocytes to adhere to vessel walls. These adhesion molecules and the molecules they bind fall
into well-defined groups (selectins, integrins).
T-lymphocytes Undergo blast transformation if stimulated by antigens, occurs in most infections. By releasing cytokines
such as IFNλ, T cells greatly increase the activity of macrophages.
Clotting system Intimately bound up with complement and kinins because of several shared activation steps.
Blood clotting is a critical part of the healing process
Fibrin The end product of blood clotting and in tissues, the matrix into which fibroblasts migrate to initiate healing.
Fibroblast Tissue cell that migrates into the fibrin clot and secretes collagen.

Modified with permission from Playfair JHL and Chaim BM: Immunology at a Glance, ed. 7, Malden, MA, Blackwell Publishing, 2001.
386 NUCLEAR MEDICINE: THE REQUISITES

and chemical mediators to the site of damage. Infection
implies the presence of microorganisms. Although infec-
tion is almost always associated with inflammation, the
reverse is not always true. The inflammatory reaction is
triggered by the products of tissue injury which can also
result from trauma, foreign particles, ischemia, and neo-
plasm. Infection can be present without inflammation, as
in severely immunosuppressed patients.
The inflammatory response results in regionally
increased blood flow, increased vascular permeability,
and emigration of leukocytes out of the blood vessels
into the tissues (chemotaxis). The plasma carries vari-
ous proteins, antibodies, and chemical mediators that
modulate the inflammatory response at the site of infec-
tion ( Fig. 12-1, Table 12-1).
RADIOPHARMACEUTICALS
Gallium-67 Citrate
Ga-67 was developed initially as a bone-seeking radio-
pharmaceutical, found use as a tumor-imaging agent,
and later was found to have infection-seeking proper-
ties. In the 1970s, Ga-67 was the mainstay of infection
scintigraphy for over a decade. Although subsequently
superceded by radiolabeled leukocytes in many clinical
settings, Ga-67 still has important utility in selected clini-
cal situations.
Chemistry and Physics
Gallium is a group III element in the Periodic Table (see
Fig. 1-1) with atomic structure and biological behavior
similar to iron ( ferric ion). The radionuclide Ga-67 is
cyclotron produced. It decays by electron capture,
emits a spectrum of gamma rays (93, 185, 288, 394 keV),
and has a physical half-life of 78 hours ( Table 12-2).
Mechanism of Uptake
The radiopharmaceutical Ga-67 citrate circulates in plasma
bound to the protein transferrin. The Ga-67 transferrin
complex is transported to the inflammatory site because
of locally increased blood flow and vascular permeability
(Box 12-2). Localization at the site of infection is to a large
extent secondary to its ferric ion-like properties.
Sites of inflammation contain iron-binding compounds
(e.g., lactoferrin released by leukocytes) and siderophores
elaborated by bacteria. After migration to an inflamma-
tory site, neutrophils release large amounts of lactoferrin.
Ga-67, with a higher binding affinity for lactoferrin
than transferrin, localizes at the site of inflammation by

Figure 12-1 Pathophysiology of inflammation. This simplified schema illustrates the body’s
response to tissue injury or infection. Permeability of the vascular endothelium plays a central role
in allowing blood cells and serum components access to the tissues. Antibodies and lymphocytes
amplify or focus these primary mechanisms. If inflammation persists beyond a few days,
macrophages and lymphocytes play an increasing role. See Table 12-1 for explanations of
inflammation terminology. (Modified with permission from Playfair JHL and Chaim BM:
Immunology at a Glance, ed. 7, Malden, MA, Blackwell Publishing, 2001.)
 Infection and Inflammation 387

dissociating from transferrin and binding to lactoferrin. proteins. It becomes firmly bound at the site of infection
There is lesser Ga-67 uptake by bacterial siderophores. by 12–24 hours. Of the injected dose, 15–25% is
excreted via the kidneys by 24 hours. Although some
Pharmacokinetics and Distribution further renal excretion occurs, the colon then becomes
Ga-67 clears slowly from the blood pool. At 24, 48, and the major route of excretion. Total body clearance is
72 hours after injection, approximately 20%, 10%, and slow, with a biological half-life of approximately 25 days.
5%, respectively, of the tracer is still bound to plasma The distribution of Ga-67 is widespread ( Table 12-3).
Soft-tissue uptake outlines the body. Liver has by far the
highest uptake, perhaps because of its metabolism of
Table 12-2 Physical Characteristics of Gallium-67 transferrin and lactoferrin ( Figs. 12-2 and 12-3). Uptake
(Ga-67), Indium-111, (In-111) and
is normal in the bone marrow and bone, and to a more
Technetium-99m (Tc-99m)
variable extent in the spleen, salivary, and lacrimal
glands. Lacrimal glands, salivary glands, and the breasts
Relative all elaborate lactoferrin. Inflammatory or stimulatory
Half-life Photopeak abundance
Radionuclide (hr) (keV ) of photons (%) processes in these organs increase production and result
in increased Ga-67 uptake, (e.g., increased uptake is seen
Ga-67 78 93 41 in the salivary glands in Sjögren’s syndrome, the lacrimal
185 23
glands in sarcoidosis [Fig. 12-4A], and the breast during
288 18
394 4 lactation [Fig. 12-5]).
Ga-67 distribution can be altered by an excess of carrier
In-111 67 173 89
247 94
gallium and whole-body irradiation. Excess ferric ion
from multiple blood transfusions and recent gadolinium
Tc-99m 6 140 89
exposure, a MRI contrast agent, also may alter biodistribu-
tion by saturation of the protein-binding sites. The result-
ing Ga-67 scan may look more like a bone scan (Fig. 12-6).
Box 12-2 Mechanisms of Localization
of Infection-Seeking Imaging Characteristics
Radiopharmaceuticals Ga-67 is not an optimal radionuclide imaging agent. It
emits four photopeaks ranging from approximately 100 to
400 keV, all with relatively low abundance ( Table 12-2).
Radiopharmaceutical Mechanism
The lower-energy photons result in a high percentage of
Gallium-67 citrate Vascular permeability, scatter relative to usable photons. The higher-energy pho-
binding to lactoferrin tons are difficult to collimate and are not efficiently
Leukocytes Diapedesis and chemotaxis detected by present-day thin gamma camera crystals. To
Nonspecific IgG Increased vascularity, maximize detection, the three lower photopeaks ( 93,
antibodies nonimmunological 185, and 300 keV ) are usually acquired.
Antigranulocyte Antibody-antigen binding
monoclonal to activated leukocytes
Methodology
antibodies
Chemotactic peptides Binding to activated
Box 12-3 describes a standard Ga-67 infection imaging
leukocytes protocol.
Tc-99m ciprofloxacin Binds to living bacteria Patient preparation with laxatives and enemas has been
Liposomes Increased vascular recommended to facilitate more rapid bowel clearance.
permeability Evidence for its effectiveness is limited. In many clinics,

Table 12-3 Normal Distribution of Radiopharmaceuticals Used for Infection Imaging

Gastro- Genito-
Radiopharmaceutical Liver Spleen Marrow Bone intestinal urinary Lung

Gallium-67 citrate *** * * * ***

Indium-111 oxine leukocytes ** *** ** *
Technetium-99m HMPAO leukocytes ** *** ** ** ** *
Antigranulocyte monoclonal antibodies ** * *** **
Nonspecific IgG antibodies *** ** ** * * *
388 NUCLEAR MEDICINE: THE REQUISITES
Figure 12-2 Normal gallium-67 distribution (male). Imaging
at 48 hours after injection shows highest uptake in the liver,
followed by bone and bone marrow. Lesser uptake is seen in the
spleen, scrotum and nasopharyngeal region. Mild uptake is
noted in the kidneys, and some intestinal clearance is noted.
Inflammatory arthritic changes are present in the knees.
they are prescribed as needed. Vigorous bowel cleansing
has been reported to cause mucosal irritation and inflam-
mation, which may result in increased Ga-67 uptake.
The usual administered adult dose for planar imaging
is 185 MBq ( 5 mCi), although sometimes higher doses of
278 MBq ( 7.5 mCi) are administered if single-photon
emission computed tomography (SPECT ) imaging is
anticipated. This dose is lower than that typically used
for tumor imaging (10 mCi [370 MBq]), where the radia-
tion dose is of lesser concern.
Whole-body imaging or limited-spot imaging may be
done as clinically indicated. A medium-energy collimator
is standard. Images are usually acquired at 48 hours,
which allows time for background clearance and an
improved target-to-background ratio. Imaging at 24 hours
may be useful in patients with suspected abscess where
early diagnosis allows for prompt intervention. Imaging
the abdomen at 24 hours can also help in differentiating
physiological bowel clearance from infection. Abdominal
activity not seen at 24 hours but seen at 48 hours likely is
due to normal bowel clearance and not a site of acute
infection. However, 24-hour images are not standard and
the usual approach, when this question arises, is to give
laxatives and obtain delayed imaging at 72–96 hours.
Normal Scintigraphy
Diffuse lung uptake of moderate degree is often present
at 24 hours but usually clears by 48 hours. The kidneys
and bladder are seen during the first 24 hours after tracer
injection owing to normal renal clearance. Subsequently,
biological clearance is through the large bowel. By 48–72
hours, the kidneys are normally only faintly visualized.
The organ with the greatest Ga-67 uptake at 48-hour
imaging is the liver. Lesser uptake is seen in the spleen.
Bone and bone marrow can be seen throughout the axial
and proximal appendicular skeleton ( see Figs. 12-2 and
12-3). Other normal sites of more variable uptake are the
nasopharynx, the lacrimal and salivary glands, and the
breast. The latter depends to some extent on the phase of
the woman’s hormonal cycle (see Fig. 12-3) and is particu-
larly prominent postpartum (see Fig. 12-5). Thymus uptake
is normal in children ( Figs. 12-7 and 12-8). The scrotum,
testes, and female perineum may have some uptake.
Postoperative sites may have Ga-67 uptake for 2–3
weeks. Ga-67 can be seen in sterile abscesses associated
with frequent intramuscular injections ( e.g., insulin
injection sites) and iron-depot injections. Increased sali-
vary gland uptake occurs after local external beam irradi-
ation or chemotherapy.
Dosimetry
The highest radiation absorbed dose ( rad) from Ga-67 is
to the large intestine, 3.7 cGy/185 MBq ( 3.7 rads/5 mCi).
Slow transit accounts for the high radiation dose. The mar-
row receives 3.5 cGy (rads) and the liver 2.2 cGy (2.2
rads). The whole-body dose is 2.2 cGy/185 MBq
(2.2 rads/5 mCi) ( Table 12-4).
Radiolabeled Leukocytes
Radiolabeled leukocytes have been used for three
decades for detection of infection and inflammation.
Both In-111 oxine and Tc-99m HMPAO labeled white
blood cells are in current clinical use.
Leukocyte Physiology
Leukocytes are the major cellular components of the
inflammatory and immune response that protect against
infection and neoplasia and assist in the repair of dam-
aged tissue. Nucleated precursor cells differentiate into
mature cells within the bone marrow. The normal blood
leukocyte count of 4.5–11.0 × 106 cells/mm3 includes
granulocytes (neutrophils, 55–65%; eosinophils, 3%; and

Figure 12-3 Normal gallium-67 distribution (female). Distribution at 48 hours is similar to the
male in Fig. 12-2, except for the addition of breast uptake. In this case, there is also more prominent
but normal lacrimal uptake and bowel activity.
basophils, 0.5%), lymphocytes (25–35%) and monocytes
(3–7%). Leukocytes spend a small part of their short
lifespan in the peripheral blood, using it mainly for trans-
portation to sites of need.
Neutrophils circulate in the peripheral blood for 5–9
hours. They respond to an acute inflammatory stimulus
by migrating toward an attractant (chemotaxis) and
enter tissues between postcapillary endothelial cells (dia-
pedesis) (see Fig. 12-1). They phagocytose the infectious
agent or foreign body and enzymatically destroy it within
cytoplasmic vacuoles. These actions are inhibited by
exposure to corticosteroids or ethanol. Leukocytes sur-
vive in tissues for only 2–3 days.
At any one time, only 2–3% of neutrophils reside in the
circulating blood. The rest are distributed in a “mar-
ginated” pool that is adherent to vascular endothelial cells
in tissues: 90% are in the bone marrow, the rest are in the
spleen, liver, lung, and, to a lesser extent, the gastroin-
testinal tract and oropharynx. These marginated cells
Infection and Inflammation 389
can be marshaled into the circulating pool by various
stimuli, including exercise, epinephrine, or exposure to
bacterial endotoxin. Basophils, also granulocytes, release
histamine, serotonin, and leukotrienes in inflammation,
and they are involved in allergic responses. Eosinophils
are particularly increased with parasitic infections.
Lymphocytes arrive at inflammatory sites during the
chronic phase of an inflammatory response. Monocytes
act as tissue scavengers, phagocytosing damaged cells
and bacteria and detoxifying chemicals and toxins. At
sites of inflammation, monocytes transform into tissue
macrophages.
Indium-111 Oxine Leukocytes
In 1976, McAfee and Thakur first radiolabeled leuko-
cytes with In-111 8-hydroxyquinoline (oxine). Many
investigations over the years have confirmed the radio-
pharmaceutical’s clinical utility. The scintigraphic
images reflect the distribution of white blood cells
390 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-4 Sarcoidosis. A, Panda sign. Prominent increased uptake in the lacrimal, parotid and
submandibular salivary glands. Nasal uptake is also prominent. B, Lambda sign. Paratracheal and
hilar nodal uptake seen with early active sarcoidosis. C, Parenchymal pulmonary uptake. Upper
lobes diffusely involved.

within the body, as well as localization at the site of
infection or inflammation.
Indium is a group III element in the Periodic Table
(see Fig. 1-1). The radionuclide In-111 is cyclotron pro-
duced. It decays by electron capture, emits two gamma
photons of 173 and 247 keV ( see Table 12-2), and has
a physical half-life of 67 hours (2.8 days).
Oxine (8-hydroxyquinolone) is a lipid-soluble complex
that chelates metal ions such as indium. As described in
the next section, labeling leukocytes with oxine cannot be
done in plasma. Alternatives to oxine have been devel-
oped that allow for labeling in plasma (e.g., tropolone and
mercaptopyridine-N-oxide [MERC]), but they are not
approved by the Food and Drug Administration ( FDA).
Fifty mL of venous blood ensures sufficient numbers
of radiolabeled leukocytes, although adequate labeling
can often be accomplished with 20–30 mL, which is
important for children. Careful handling is required to
avoid damaging the cells, which otherwise might result
in loss of their ability to migrate or even their viability.
Proper labeling does not adversely affect normal physio-
logical function, and the radiolabel usually remains stable
in vivo for over 24 hours.
The higher the serum leukocyte count, the more
likely that labeling efficiency will be high. Ideally the
patient’s leukocyte count should be >5000/mm3,
although diagnostic scintigraphy can often be performed
with patient counts as low as 3000/mm3.
Because In-111 is cyclotron produced, it usually must
be ordered the day prior to cell labeling. Radiolabeling
should be performed in a well-equipped laboratory with
a laminar flow hood. The in vitro labeling procedure

Figure 12-5 Postpartum gallium-67 breast uptake. The
increased production of lactoferrin results in prominent
postpartum breast uptake. Although some uptake is normal in the
non-lactating female ( see Fig. 12-3), it is considerably less and
quite variable, dependent on the normal hormonal cycle. Two
intensity settings are shown.
requires a minimum of 2 hours. Lacking facilities and
trained personnel, most hospitals send the patient’s
blood to an outside commercial pharmacy for radiola-
beling. Thus, it may be 3–4 hours before the cells are
reinfused. The longer the interval between blood with-
drawal and reinfusion, the higher the likelihood is that
the cells will lose viability.
Details of the radiolabeling process are described in
Box 12-4. The patient’s withdrawn blood is allowed to
settle, and the majority of the erythrocytes are removed.
In-111 oxine indiscriminately labels granulocytes, lym-
phocytes, monocytes, platelets, erythrocytes, and also
plasma transferrin. Thus the plasma must be separated
and removed from the cells by centrifugation to ensure
preferential cell labeling. It is kept for later resuspen-
sion of the leukocytes before reinfusion.
The leukocyte pellet is suspended in normal saline
and incubated with In-111 oxine. The lipid solubility of
the In-111 oxine complex allows it to diffuse through
cell membranes. Intracellularly, the complex dissoci-
ates. Oxine diffuses back out of the cell, whereas In-111
binds to nuclear and cytoplasmic proteins.
Infection and Inflammation 391
Figure 12-6 Abnormal gallium-67 distribution. This patient
received chemotherapy the day before administration, an unusual
situation and likely the cause of these findings. Almost no uptake is
seen in the liver, the organ normally having greatest Ga-67 uptake.
This has also been reported in patients with multiple blood
transfusions or recent gadolinium contrast administration with MRI.
Pure granulocyte preparations have been radiolabeled;
however, they require elaborate density gradient separa-
tion techniques and have not shown a clear clinical
advantage. Thus, they are not generally used.
Standard quality control measures (e.g., testing for
sterility and pyrogenicity) cannot be performed because
of the need for prompt reinfusion after labeling to ensure
cell viability. However, the final radiopharmaceutical
preparation is examined for abnormal morphology,
clumping, and excessive red cell contamination. The final
preparation contains predominantly radiolabeled granulo-
cytes, lymphocytes, and monocytes, but there will also be
10–20% platelets and erythrocytes. The percent of label-
ing efficiency is routinely determined ( see Box 12-3).
Typical labeling efficiency ranges from 75–90%. When it
is <50%, the cells should probably not be reinfused.
The ultimate test of viability of leukocytes is their
in vivo function manifested by a normal distribution
392 NUCLEAR MEDICINE: THE REQUISITES
Box 12-3 Gallium-67 Citrate Imaging:
Protocol Summary
PATIENT PREPARATION
No recent barium contrast studies
RADIOPHARMACEUTICAL
Gallium-67 citrate, 185 MBq (5 mCi) injected
intravenously
INSTRUMENTATION
Camera: Large-field-of-view gamma camera
Photopeak: 20% window over 93 keV, 185 keV, and
300-keV photopeaks
Collimator: Medium energy
IMAGING PROCEDURE
24-hr images (optional): Site of suspected infection if
early intervention considered; abdominal images may
be helpful for interpreting activity seen at 48 hours
48-hr images: Whole body imaging, including head and
extremities, unless the site of suspected infection is
limited to one site, e.g., hip prosthesis
Delayed 72- to 96-hr images of abdomen as indicated
to differentiate intraabdominal infection from normal
bowel clearance; laxatives or enemas as needed
SPECT of the abdomen, pelvis, or chest as indicated
Figure 12-7 Thymus uptake of Ga-67. This child’s malignant
lymphoma responded to chemotherapy. On this follow-up Ga-67
study, the coronal SPECT cross-sectional slice shows thymus
uptake. Thymus uptake may be seen normally in children.
within the body and their ability to detect infection. If
the infused leukocytes become nonviable, as might
result from an interval greater than 4 hours between
labeling and reinfusion, increased liver and lung
uptake is typically seen on scintigraphy. With exces-
sive erythrocyte and platelet labeling, blood pool is
prominent.
After infusion of the radiolabeled leukocytes, no sig-
nificant elution of the In-111 from the leukocytes occurs.
The effective half-life of clearance from the blood circula-
tion is 7.5 hours.
Initial distribution after reinfusion is seen in blood pool,
lungs, liver, and spleen. Lung activity is the result of cellu-
lar activation from in vitro cell manipulation. By 4 hours
after reinjection, lung activity and blood pool normally
decrease, although not always completely ( Fig. 12-9). By
24 hours, blood-pool activity is not normally present.
Persistent blood pool at 24 hours indicates a high per-
centage of labeled erythrocytes or platelets.
At 24 hours, the most intense uptake is seen in the
spleen, followed by the liver and then the bone marrow
( see Fig. 12-9). Neither genitourinary nor bowel activity
is normally seen. Table 12-3 compares this normal organ
distribution of radiolabeled leukocytes with other infec-
tion-seeking scintigraphic agents.
The spleen receives the highest radiation-absorbed
dose, approximately 15–20 cGy/18.5 MBq (15–20
rads/500 μCi) in adults, but up to 30–50 cGy (rads) in
small children ( Table 12-4). In addition to gamma emis-
sions, In-111 emits low-energy conversion and Auger
electrons of 0.6–25.4 keV with a short range in tissue
that can damage labeled cells. Neutrophils are not typi-
cally damaged; however, because lymphocytes are more
radiosensitive and longer-lived, there is a theoretical con-
cern for possible mutagenic and oncogenic effects. Data
are limited. The high splenic radiation dose is the reason
In-111 leukocytes are rarely used in children.
Imaging Methodology
A typical imaging protocol for In-111 oxine leukocytes is
detailed in Box 12-5. Both the 173- and 247-keV photo-
peaks are acquired. A medium-energy collimator is used.
Whole-body imaging is routine, although high-count spot
images and SPECT are acquired when needed.
Images are routinely acquired 18–24 hours after radio-
pharmaceutical injection ( Box 12-6). This allows suffi-
cient time for leukocyte localization and blood pool
clearance. Further delayed images do not often give addi-
tional information. Earlier imaging ( e.g., at 4 hours) is
less sensitive than 24 hours for detecting infection but
may occasionally be useful for urgent diagnosis (e.g., an
abscess that requires prompt intervention).
Four-hour imaging is mandatory for the localization of
inflammatory bowel disease because the inflamed
mucosal cells slough, become intraluminal, and move
distally. Twenty-four hour images may provide mislead-
ing and erroneous information as to the site of inflamma-
tion. Fixed activity between 4- and 24-hour images
would suggest an abscess as a complication of the inflam-
matory bowel disease.
In some cases, diagnostic accuracy is improved by
performing an additional scintigraphic study for direct
correlation using a different radionuclide with a different
mechanism of uptake. In current practice, the most
common one is a Tc-99m sulfur colloid ( Tc-99m SC)
 Infection and Inflammation 393

Figure 12-8 Gallium-67 uptake in heart in myocarditis and thymus. A 20-month-old child
received azathioprine and steroids for treatment of idiopathic myocarditis. A, Left, Pre-therapy
planar image of the chest showed no abnormal uptake. Right, Post-therapy planar image shows
prominent uptake by the thymus (arrowhead). B, In contrast to the planar study, the pretherapy
SPECT study showed myocardial uptake (best seen on middle image). Three sequential transverse
slices through the myocardium are shown.

bone-marrow scan. Dual-isotope studies require special

Table 12-4 Radiation Dosimetry for Ga-67 Citrate, attention to the imaging characteristics of the radionu-
In-111 Oxine and Tc-99m HMPAO clides (e.g., photopeaks, half-lives, downscatter, the rela-
Leukocytes (WBCs) tive administered doses, and the camera’s capability for
simultaneous multichannel acquisition).
In-111 oxine Tc-99m The problem of downscatter or upscatter should be
WBCs HMPAO considered. One approach is to perform the Tc-99m scan
Ga-67 citrate cGy/18.5 MBq WBCs
cGy/185 MBq or cGy/370 MBq first because of its shorter half-life. By 24 hours, 94% of
Organ or rad/5 mCi rad/500 μCi or rad/10 mCi the Tc-99m activity will have decayed. Blood required for
In-111 leukocyte labeling could be drawn prior to injec-
Bladder wall 2.8
tion of the Tc-99m tracer and the In-111 radiolabeled cells
Large intestine 3.7 3.6
Liver 2.2 2.66 1.5 reinfused after acquiring the Tc-99m scan. In-111 leuko-
Bone marrow 3.5 1.99 1.6 cyte scintigraphy would be acquired the next day.
Spleen 1.8 20.00 2.2 An alternative approach is to acquire both studies
Ovaries 1.5 0.20 0.3 simultaneously, using a dual-isotope acquisition tech-
Testes 1.0 0.014 1.9
nique. This approach ensures identically positioned
Total body 2.2 0.37 0.3
images for direct comparison. One approach to mini-
WBCs, White blood cells. mize upscatter from Tc-99m is to acquire only the upper
Target organ dose is in bold. 247-keV photopeak of In-111. Alternatively, one could
394 NUCLEAR MEDICINE: THE REQUISITES
Box 12-4 Radiolabeling Autologous
Leukocytes with Indium-111
Oxine
PREPARATION
Patient’s peripheral leukocyte count should be greater
than 5000 cells/mm3.
PROCEDURE
1. Collect autologous blood:
Draw 30 to 50 ml into an ACD anticoagulated
syringe using a 19-gauge needle.
2. Isolate leukocytes:
Separate red blood cells (RBCs) by gravity
sedimentation and 6% Hetastarch, a settling
agent.
Centrifuge the leukocyte-rich plasma (LRP) at 300
to 350 g for 5 min to remove platelets and
proteins.
A white blood cell (WBC) button forms at the
bottom of the tube.
Draw off and save the leukocyte-poor plasma (LPP)
for later washing and resuspension.
3. Label leukocytes:
Suspend WBCs (LRP) in saline (includes
granulocytes, lymphocytes, monocytes, and some
RBCs).
Incubate with In-111 oxine for 30 min at room
temperature and gently agitate.
Remove unbound In-111 by centrifugation. Save
wash for later calculation of labeling efficiency.
4. Prepare injectate:
Resuspend 500 μmCi In-111 leukocytes in saved
plasma (LPP).
Inject via peripheral vein within 2 to 4 hr.
5. Perform quality control:
Microscopic examination of cells.
Calculate labeling efficiency:
Assay the cells and wash in dose calibrator.
Labeling efficiency = C/([C + W] × 100%)
C, activity associated with the cells; W, activity
associated with the wash.
acquire a narrow window for the lower 173-keV photo-
peak. Downscatter of In-111 into the Tc-99m window is
not a significant problem because of the low activity of
In-111 (500 μCi) used compared to Tc-99m (20 mCi).
Image Interpretation
Activity outside the expected normal distribution of
leukocytes suggests infection ( Figs. 12-10 to 12-16).
Focal uptake equal to or greater than that of the liver is
typical for an abscess. Activity equal to the liver gener-
ally signifies a clinically important inflammatory site;
activity less than bone marrow suggests a low-level
inflammatory response.
One should keep in mind interpretive pitfalls and
potential false-positive and false-negative findings with
leukocyte scintigraphy (Box 12-7). Leukocytes may accu-
mulate at sites of inflammation without clinical infection
(e.g., intravenous catheters; nasogastric, endogastric,
and drainage tubes; tracheostomies; colostomies; and
ileostomies). Unless very intense, this uptake should not
be considered abnormal. Uninfected postsurgical wounds
commonly show faint uptake for up to 2 weeks. If uptake
is intense, persists, or extends beyond the surgical wound
site, infection should be suspected ( Fig. 12-12). Low-
grade uptake may be seen at healing bone-fracture sites.
Activity seen within the abdomen may be due to
a variety of causes not due to infection and thus have the
potential for misinterpretation. These include pseudo-
aneurysm, noninfected hematomas, and accessory
spleens ( Fig. 12-17). Renal transplants normally accu-
mulate radiolabeled leukocytes.
Even intraluminal intestinal activity can be the result
of swallowed or shedding cells that occur with herpes
esophagitis, pharyngitis, sinusitis, pneumonia, or gastro-
intestinal bleeding ( Fig. 12-18). Very rarely, tumors have
uptake.
Technetium-99m HMPAO Labeled Leukocytes
Leukocytes labeled with Tc-99m have a number of advan-
tages over In-111 leukocytes ( Table 12-5). Being genera-
tor produced, Tc-99m is available when needed. The
radiation absorbed dose to the patient is much lower,
which is particularly important for pediatric patients.
The greater activity that can be administered results in
a higher photon yield and better image quality. The more
optimal Tc-99m photopeak results in superior image qual-
ity. This all potentially translates into improved infection
detectability.
The element technetium is in group VIIB of the
Periodic Table (see Fig. 1-1). The radionuclide Tc-99m is
generator-produced from molybdenum-99. It decays by
isomeric transition, emits one gamma photon of 140 keV
(see Table 12-1), and has a physical half-life of 6 hours.
Tc-99m hexamethyl-propylene-amine oxime ( Tc 99m
HMPAO; Ceretec, Mediphysics) was initially approved by
the FDA for cerebral perfusion imaging. HMPAO
is lipophilic, allowing it to cross cell membranes.
In the cerebral cortex, after crossing the blood–brain
barrier, it is taken up intracellularly in cortical
cells. Intracellularly, Tc-99m HMPAO changes into
a hydrophilic complex and becomes trapped, bound to the
mitochondria and the nucleus. This observation led to the
development of a labeling technique for leukocytes using
Tc-99m HMPAO.
With Tc-99m HMPAO labeled leukocytes, the colon
is the organ receiving the highest radiation dose
(3.6 cGy/370 MBq [3.6 rads/10 mCi]), followed by the
bladder and spleen ( 2.2 cGy/MBq [2.2 rads/10 mCi])
 Infection and Inflammation 395

Figure 12-9 Normal distribution of indium-111 oxine leukocytes at 4 and 24 hours. Anterior and
posterior whole-body images. The highest uptake is seen in the spleen, followed by the liver, then
the bone marrow. No intestinal or renal activity is present which is normal. The 4-hour images
show some lung uptake that resolves by 24 hours. There is no other apparent change in distribution
between 4 and 24 hours.

( see Table 12-4). Unlike In-111, there is no concern
regarding potential direct radiation damage to leukocytes.
The methodology for leukocyte radiolabeling with
Tc-99m HMPAO was described in 1986 and is very simi-
lar to that used for labeling leukocytes with In-111
oxine. In contrast to In-111 oxine, Tc-99m HMPAO
leukocyte labeling can be performed in plasma. HMPAO
preferentially labels granulocytes, a potential advantage
for imaging acute purulent processes. The radiolabeling
process does not adversely affect leukocyte function.
The FDA views Tc-99m HMPAO–labeled leukocytes as
an acceptable alternative use of an approved radiophar-
maceutical.
The biological half-life of Tc-99m HMPAO labeled
leukocytes in blood is shorter than that of In-111 oxine
leukocytes (4 hours versus 6 hours) due to the slow elu-
tion of the Tc-99m HMPAO from circulating labeled cells.
Tc-99m HMPAO–labeled leukocytes distribute in the
body similar to In-111 oxine leukocytes, with localiza-
tion in the spleen, kidney, and bone marrow ( Fig. 12-19).
Early lung uptake similar to that seen with In-111 oxine
occurs but also decreases by 4 hours. Unlike In-111-
oxine–labeled cells, Tc-99m HMPAO leukocytes are par-
tially cleared by the hepatobiliary and renal systems with
excretion of a secondary hydrophilic complex, which
also occurs with Tc-99m HMPAO cerebral perfusion
396 NUCLEAR MEDICINE: THE REQUISITES

Box 12-5 Indium-111 Oxine Leukocyte

Scintigraphy: Protocol Summary

RADIOPHARMACEUTICAL
In-111 oxine in vitro labeled leukocytes, 500 μCi
(18.5 MBq)

INSTRUMENTATION
Camera: Large field of view
Windows: 20% centered over 173 and 247 keV
photopeaks
Collimator: Medium energy

PATIENT PREPARATION
Draw 50 ml of blood. Radiolabel cells in vitro (Box 12-3)

PROCEDURE
Inject labeled cells intravenously, preferably by direct
venipuncture through a 19-gauge needle. Contact
with dextrose in water solutions may cause cell
damage.
Imaging at 4 hr may be helpful to diagnose an acute
abscess and is critical in localizing inflammatory
bowel disease.
Perform routine whole body imaging at 24 hr.
Acquire anterior abdomen for 500k counts, then other
images for equal time. Include anterior and posterior
views of the chest, abdomen, and pelvis, and spot
images of specific areas of interest (e.g., feet) for a
minimum of 200k counts or 20 min.
Perform SPECT in selected cases.

Box 12-6 Optimal Imaging Time for

Infection-Seeking
Radiopharmaceuticals

Radiopharmaceutical Time (hr)

Gallium-67 48
Indium-111 leukocytes 24
Nonspecific IgG antibodies 10–24 Figure 12-10 Liver abscess: In-111 oxine leukocytes.
Transverse (above) and coronal (below) cross-sectional SPECT
Antigranulocyte monoclonal antibodies 1–6
slices with focal uptake of leukocytes in the right lobe of the liver.
Technetium-99m HMPAO leukocytes 1–4 An abscess was subsequently drained.
Chemotactic peptides 1–4
Technetium-99m nanocolloids 1
Fluorine-18 fluorodeoxyglucose 1
Imaging Methodology
The imaging methodology is somewhat different for
Tc-99m HMPAO than for In-111 oxine labeled leukocytes
imaging. The kidneys and bladder may be seen by 1–2 because of its shorter physical half-life and its hepatobil-
hours after injection. The gallbladder is visualized in 4% iary and urinary clearance. A detailed protocol is
of patients at 1 hour and in about 10% by 24 hours. described (Box 12-8). Imaging of the abdomen should be
Biliary clearance may be seen as early as 2 hours and performed between 1 and 2 hours after reinfusion in order
bowel activity is routinely visualized by 3–4 hours and to avoid hepatobiliary and urinary clearance (see Box 12-5).
increase with time. Four-hour imaging is preferable for peripheral extremities.

Figure 12-11 Intra-abdominal abscess: In-111 oxine
leukocytes. Anterior view of pelvis. Focal uptake in the right lower
quadrant caused by a perforated appendix with abscess formation.
Figure 12-12 Postoperative wound infection: In-111 oxine leukocytes. Dehiscence of the incision
site because of an abscess inferior and deep to incision. Note the more intense uptake inferiorly.
Infection and Inflammation 397
Considerable blood-pool activity is commonly seen
because of the early imaging period of Tc-99m HMPAO
leukocytes compared to In-111 oxine leukocytes, which
may complicate interpretation, particularly in the chest.
The high count rate of Tc-99m allows for better SPECT
quality than In-111. Delayed imaging up to 24 hours may
be useful on occasion ( Fig. 12-20).
Image Interpretation
The discussion under In-111 oxine leukocytes applies
here. The only difference is the biliary and genitourinary
excretion of the Tc-99m HMPAO. Uptake outside that
expected for normal distribution suggests infection
( Fig. 12-21). Relative advantages of Tc-99m HMPAO ver-
sus In-111 oxine leukocytes are listed in Table 12-5. These
will be discussed further.
Technetium-99m Fanolesomab ( NeutroSpec)
Alternative infection-seeking radiopharmaceuticals have
been sought that do not require cell labeling and would
not have the potential for transmission of bloodborne dis-
eases. An antigranulocyte monoclonal antibody, Tc-99m
fanolesomab ( NeutroSpec, Mallinckrodt), was approved
by the FDA in 2004 for scintigraphy in patients with
equivocal signs and symptoms of acute appendicitis.
398 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-13 Infected aortofemoral graft: In-111 leukocytes. Left, Anterior abdomen. Right,
Anterior pelvis. Abnormal uptake confirms the clinical suspicion of a surgical graft infection
(arrowheads).

Figure 12-14 Infected thoracic aortic graft: In-111 leukocytes. A, Postoperative chest
radiograph. B, In-111 oxine leukocytes localize in the aortic graft in the region of the aortic knob.

Tc-99m fanolesomab is a murine IgM monoclonal
antibody that binds avidly to surface CD-15 antigens
expressed on human neutrophils, monocytes, and
eosinophils. Less than 5% of circulating leukocytes are
monocytes or eosinophils. Therefore, uptake is prima-
rily due to neutrophils.
Blood concentrations decrease rapidly with an initial
half-life of 0.3 hours and a second phase half-life of
8 hours. At 2 hours postinjection, the liver has the highest
radioactivity with 50% of the total injected dose, followed
by the kidney, spleen, and red marrow. At 26–33 hours
after injection, 38% is cleared through the urinary tract.
The spleen receives the highest radiation absorbed
dose (2.3 rads/10 mCi or 2.3 cGy/370 MBq) (Table 12-6),
approximately in the range of Tc-99m HMPAO leuko-
cytes. The liver and bladder receive the next highest
radiation dose.
As with all murine derived antibodies, autoimmume
HAMA reactions are a concern. The incidence of human
antimouse antibodies is quite low after a single adminis-
tration (near zero), although detection of antibodies is
dependent on the sensitivity of the assay. Limited data
suggest that repeat injections result in a slightly higher
incidence of about HAMA 10% but have a very low inci-
dence of clinical side effects.
The usual adult administered dose is 10–20 mCi
(370–740 mBq). In investigational studies, dynamic
imaging was acquired, starting at the time of injection.
Initially, ten 4-minute images were acquired. The
patient voids and static images are acquired, including
 Infection and Inflammation 399

supine anterior, posterior, right anterior oblique and left

Figure 12-15 Pneumonia: In-111 leukocytes. Posterior chest
view. Focal uptake in the left lower lobe. The purpose of the study
was to locate the source of postoperative fever. Pneumonia was
not suspected clinically. The last chest x-ray had been 10 days
earlier. A subsequent radiograph confirmed the diagnosis.
anterior oblique views, followed by a standing anterior
image if the concern for infection is in the abdomen or
pelvis. Finally, a static one-million count image is
obtained. Imaging up to 90 minutes is recommended.
This protocol was specific for acute appendicitis.
Image Interpretation
Normal distribution includes the blood pool, liver,
spleen, bone marrow, and urinary excretion. A diagnos-
tic abnormality is characterized by the presence of focal
irregular, asymmetric uptake of radiotracer localized at
the site of inflammation or infection ( Fig. 12-22). The
abnormal localization remains constant or increases in
intensity in follow-up imaging.
Clinical Utility and Accuracy
See later section on Intra-Abdominal Infection and
Table 12-9.
Technetium-99m Sulesomab ( LeukoScan)
This radiopharmaceutical is commercially available in
Europe. LeukoScan ( Immunomedics, Morris Plains,
N.J.), is Tc-99m-labeled antigranulocyte ( IgG1) murine

Figure 12-16 Osteomyelitis of the right maxillary sinus: In-111 leukocytes. History of bilateral
sinus surgery. A, Bone scan shows fairly symmetrical ethmoid and maxillary sinus uptake. B, Indium-
111 oxine leukocyte study shows uptake just right of midline in a pattern different from the bone
scan, consistent with focal maxillary sinus infection, abscess, or osteomyelitis. Osteomyelitis was
confirmed at surgery.
400 NUCLEAR MEDICINE: THE REQUISITES

Fluorine-18 Fluorodeoxyglucose
Box 12-7 Interpretative Pitfalls: False F-18 fluorodeoxyglucose ( FDG) PET imaging is used on
Negative and False Positive a clinical basis primarily for tumor imaging, and to
Leukocyte Scans a lesser extent, cardiac and brain imaging. A common
cause for abnormal uptake not due to malignancy is that
FALSE NEGATIVE of inflammation and infection, due to glucose utilization
Encapsulated, nonpyogenic abscess by activated granulocytes and macrophages.
Vertebral osteomyelitis F-18 FDG PET has potential advantages over radio-
Chronic low-grade infection labeled leukocyte studies. The study length is com-
Parasitic, mycobacterial, or fungal infections pleted within 2 hours of radiopharmaceutical injection
Intrahepatic or perihepatic or splenic infection and PET image resolution is superior to single-photon
Hyperglycemia imaging. Furthermore, the problems associated with
Steroids radiolabeling leukocytes and reinfusion of blood prod-
FALSE POSITIVE ucts is eliminated.
Preliminary investigations suggest that F-18 FDG may
Gastrointestinal bleeding
Pseudoaneurysm
be useful for diagnosing a variety of infections (e.g.,
Healing fracture inflammatory bowel disease, fever of unknown origin,
Soft tissue tumor and osteomyelitis). Although investigations have sought
Swallowed leukocytes; oropharyngeal, esophageal, or to use it to differentiate aseptic from septic hip prosthe-
lung disease ses, false positives are not uncommon. Radiolabeled
Surgical wounds, stomas, or catheter sites leukocyte imaging combined with bone marrow scintig-
Hematomas raphy is superior. For knee prostheses, the specificity is
Tumors even poorer than for the hip.
Accessory spleens
Renal transplant Investigational Radiopharmaceuticals
There continues to be active investigation for new
infection-seeking radiopharmaceuticals ( see Box 12-1).
Investigation has changed over time from developing
antibody Fab' fragment. Fab' fragments have less large proteins with nonspecific uptake mechanism ( IgG)
immunoreactivity than whole antibodies and a better tar- to receptor specific proteins of large size (antigranulo-
get-to-background ratio because of their rapid renal cyte antibodies) and moderate size (antibody fragments)
clearance. Early clinical trials have found the accuracy of to small receptor-binding proteins and peptides
LeukoScan imaging to be equal or superior to that of In- (cytokines). Some of these investigational radiopharma-
111 leukocyte imaging particularly for musculoskeletal ceuticals are briefly discussed.
infection and osteomyelitis. Imaging can be performed Nonspecific polyclonal immunoglobulin (Ig) G
within 1–6 hours after injection. shows localization in a variety of clinical infections.

Figure 12-17 False positive In-111 leukocyte scan caused by accessory spleens. A 78-year-old
woman with bacterial endocardititis had prior splenectomy. The In-111 leukocyte study was
ordered to localize extra-cardiac infection. A Tc-99m SC study confirmed that the focal uptake in the
left upper quadrant was due to accessory spleens (see Fig. 7-49).
 Infection and Inflammation 401

Figure 12-18 False positive In-111 leukocyte study. Fever of uncertain etiology. Images obtained
at 4 hours (A) and 24 hours (B). Although there is leukocyte localization predominantly to the
transverse bowel at 4 hours, it is seen in the descending colon and rectum on 24-hour images. The
patient had gastrointestinal bleeding on the day of the exam. The fever resolved spontaneously. The
intraluminal activity was presumed due to gastrointestinal bleeding.

Accumulation is not the result of an immunological Tc-99m ciprofloxacin ( Infecton) is a Tc-99m radio-
mechanism, but rather from the increased vascular per- labeled fluoroquinolone broad-spectrum antimicrobial
meability associated with inflammatory processes. The agent that binds to DNA of living bacteria. One early
gastrointestinal and genitourinary systems show varying study has shown good general sensitivity similar to radio-
degrees of uptake. Slow blood clearance requires a mul- labeled leukocyte studies. Of particular note, vertebral
tiple day imaging protocol. Preliminary data suggests infections had increased uptake in five of six patients
good accuracy; however, there is no commercial interest with negative radiolabeled leukocyte studies. This radio-
in this agent. pharmaceutical is under investigation.
Chemotactic peptides or cytokines are involved in the
initiation, amplification, and termination of the inflamma-
tory response. They act through interaction with specific
cell-surface receptors. Produced by bacteria, chemotactic
Table 12-5 Advantages/Disadvantages of In-111
peptides bind to receptors on the cell membrane of neu-
Oxine versus Tc-99m HMPAO Labeled
trophils, stimulating the cells to undergo chemotaxis.
Leukocytes
Analogs of these peptides have been synthesized and radi-
olabeled. Localization at sites of infection is rapid owing
In-111 Tc-99m to the small size of these compounds; they easily pass
oxine HMPAO
leukocytes leukocytes through vascular walls and quickly enter an abscess. High
target-to-background ratio occurs at 1 hour. Radiolabeled
Radionuclide immediately available No Yes cytokines (e.g., interleukin-8) show promise.
Stable radiolabel, no elution from cells Yes No
Liposomes are spherical envelopes of cell membrane
Allows labeling in plasma No Yes
Dosimetry Poor Good that have been investigated over the years for various
Early routine imaging No Yes indications. Newer improved techniques have revital-
Long half-life allows for delayed Yes No ized its potential for infection imaging. A new approach
imaging is to target endothelial adhesion molecules expressed
Imaging time Long Short
during inflammation (e.g., anti-E-selectin).
Permits dual isotope imaging Yes No
Bowel and renal clearance No Yes Radiolabeled lymphocytes are potentially useful for
Image resolution Good Fair diagnosing chronic and more indolent inflammatory
processes, such as rejection of kidney and heart transplants.
402 NUCLEAR MEDICINE: THE REQUISITES
Figure 12-19 Normal Tc-99m HMPAO leukocyte distribution
4 hours postinjection. Uptake is greatest in the spleen, followed
by the liver and bone marrow, similar to that seen with In-111
leukocytes in Fig. 12-9. Unlike In-111 leukocyte distribution,
Tc-99m HMPAO shows bowel and urinary clearance. Low-grade
diffuse pulmonary uptake is also seen. The study was performed
because of suspected left knee prosthesis infection.
Only preliminary studies have been reported. Unlike
neutrophils, lymphocytes are quite radiosensitive.
Concerns have been expressed about the radiation
effect on function, viability, and, more importantly, the
potential for oncogenesis because of the lymphocytes’
long lifespans.
CLINICAL APPLICATIONS FOR
INFECTION SCINTIGRAPHY
Many of the factors that need to be considered in decid-
ing which radiopharmaceutical is indicated for a specific
clinical situation have been mentioned. This section
focus on individual disease processes and discusses the
pathophysiology of some of the more common problems,
preferred radiopharmaceuticals for specific indications,
and any special methodologies that should be used.
Box 12-8 Technetium-99m HMPAO
Leukocyte Scintigraphy:
Protocol Summary
PATIENT PREPARATION
Wound dressings should always be changed prior to
imaging.
RADIOPHARMACEUTICAL
Tc-99m hexamethylpropylene amine (HMPAO) in vitro
labeled leukocytes, 10 mCi (370 MBq)
INSTRUMENTATION
Camera: Large field of view; two-headed camera
preferable for whole body imaging
Windows: 15%, centered over 140-keV photopeaks
Collimator: Low energy, high resolution
PATIENT PREPARATION
Draw 50 ml of blood to radiolabel cells in vitro
PROCEDURE
Radiolabel the patient’s leukocytes in vitro with Tc-99m
HMPAO.
Reinject labeled cells intravenously, preferably by direct
venipuncture through 19-gauge needle. Contact with
dextrose in water solutions may cause cell damage.
IMAGING
Imaging between one and two hours is mandatory for
intra-abdominal imaging or to localize inflammatory
bowel disease. Imaging at 4 hr or later may be
advantageous for peripheral skeletal imaging, e.g.,
osteomyelitis of feet.
Whole body imaging: Two-headed camera with whole
body acquisition for 30 min; 10-min spot images for
regions of special interest
SPECT in selected cases
General Considerations
In-111 leukocytes have replaced Ga-67 for most indica-
tions. Ga-67 has suboptimal imaging characteristics due
to its multiple high-energy photopeaks with low abun-
dance and high scatter fraction. Intra-abdominal clear-
ance via the bowel limits intra-abdominal diagnosis and
the usual 48-hour imaging time is clinically disadvanta-
geous. The major role for Ga-67 is for the diagnosis
of pulmonary inflammatory disease such as sarcoidosis,
pneumocystis, and drug-induced etiologies (e.g.,
bleomycin). Ga-67 may also be useful in the leukopenic
patient ( Box 12-9). It has also been successfully used for
discitis in the spine and vertebral osteomyelitis.

Generally, for acute infection, the largest and most suc-
cessful experience has been with radiolabeled leuko-
cytes. Tc-99m HMPAO has distinct advantages over
In-111 leukocytes ( see Table 12-5). Being generator pro-
duced, Tc-99m is available for same-day radiolabeling,
whereas In-111 is cyclotron-produced and must be
ordered (and in most cases radiolabeled) the following
day. Tc-99m has superior imaging characteristics and
much greater activity can be administered, resulting in
higher count images and better resolution. Imaging is
routinely performed on the same day of administration
with Tc-99m HMPAO, whereas In-111 leukocytes are
Figure 12-20 Infected arterial-venous graft imaged with Tc-
99m HMPAO leukocytes. Image obtained at 4 hours (above) shows
focal uptake within the graft on the right. Delayed image at 24
hours show increasing uptake in the same region.
Figure 12-21 Postoperative empyema: Tc-99m HMPAO leukocytes. The infection occurred
following thoracotomy for lung cancer. Left, Posterior view. Right, Anterior view.
Infection and Inflammation 403
usually imaged at 24 hours. The considerably lower radi-
ation absorbed dose to the spleen of Tc-99m HMPAO
makes it the agent of choice for pediatric patients.
Tc-99m HMPAO is cleared via hepatobiliary and geni-
tourinary excretion. Abdominal imaging with Tc-99m
HMPAO must be performed before intra-abdominal clear-
ance, at 1–2 hours. Because leukocytes may take many
hours to localize, detectability may be less at this early
time period. Thus, In-111 labeled leukocytes are the
agent of choice for intra-abdominal infection (Table 12-7),
with the only exception being pediatric patients and
inflammatory bowel disease.
Radiolabeled antigranulocyte antibodies have been
approved for infection scintigraphy in the United
States ( Tc-99m fanolesomab, NeutroSpec) and in
Europe ( Tc-99m sulesomab, LeukoScan). Although
NeutroSpec has only been approved for diagnosis of
acute appendicitis, it is anticipated that it will have
a much wider applicability because it does not require
leukocyte radiolabeling. However, at present, data is
limited for other indications.
There is concern about the sensitivity of leukocyte
imaging in patients with clinical conditions or therapies
that might alter leukocyte function, such as hyper-
glycemia, steroid therapy, chemotherapy, hemodialysis,
and hyperalimentation. Data are limited.
Conflicting data exists regarding the sensitivity of
In-111 leukocyte scintigraphy for detecting infection in
patients receiving antibiotics. The discrepant reports
may be due to the adequacy or inadequacy of treatment.
Invariably many patients who have leukocyte imaging
are on antibiotics.
There has also been concern that false-negative stud-
ies might occur with chronic infection. However, most
investigations have found no significant difference in
sensitivity for detection of acute or chronic infections.
Although chronic inflammations consist largely of lym-
phocytes, monocytes, plasma cells, and macrophages,
they also have significant neutrophilic infiltration and, at
404 NUCLEAR MEDICINE: THE REQUISITES

times, frank pus. Furthermore the In-111–labeled
mixed-cell population contains many lymphocytes.
Tc-99m HMPAO preferentially labels granulocytes.
Thus, some chronic infections may not be detected with
In-111 oxine labeled leukocytes. Ga-67 may have a role in
patients with low-grade infection (e.g., fungal, protozoa).
Osteomyelitis
The histopathology of osteomyelitis during the acute
phase includes neutrophilic inflammation, edema, and vas-
cular congestion. Because of the bone’s rigidity, intra-
medullary pressure increases, compromising the blood
supply and causing ischemia and vascular thrombosis.
Over several days, the suppurative and ischemic injury may
result in bone fragmentation into devitalized segments
called sequestra. Infection may spread via Haversian and
Table 12-6 Radiation Dosimetry for Tc-99m
Fanolesomab (NeutroSpec)
5-year-old
Adults children cGy/
cGy/370 MBq 37 MBq or
Organ or rads/10 mCi rads/mCi
Volkmann’s canals to the periosteum, resulting in
abscesses, soft-tissue infection, and sinus tracts.
With persistent infection, chronic inflammatory cells,
including lymphocytes, histiocytes, and plasma cells,
join the neutrophils. Fibroblastic proliferation and new
bone formation occur. Periosteal osteogenesis may sur-
round the inflammation to form a bony envelope, or
involucrum. Occasionally a dense fibrous capsule con-
fines the infection to a localized area of suppuration
called a Brodie’s abscess.
Bone infection is usually bacterial in origin. Micro-
organisms reach the bone by one of three mechanisms:
hematogenous spread, extension from a contiguous site
of infection, and direct introduction of organisms into
bone by trauma and surgery.
The terminology of acute and chronic osteomyelitis
is frequently confused. The word acute implies
hematogenous spread. Chronic osteomyelitis is not
chronic in the traditional sense. It is nonhematogenous
in origin, but is an active infection with a neutrophilic
inflammatory component, although with time there may
be a subacute or chronic inflammatory response as well.
Perhaps chronic active osteomyelitis would be better
terminology.

Bladder wall 1.20 0.27 Acute Hematogenous Osteomyelitis

Large intestine 0.34 0.21 In children, bone is usually infected by hematogenous
Liver 1.80 0.41 spread. It usually involves the red marrow of long bones
Bone marrow 0.38 0.11 due to its relatively slow blood flow in metaphyseal capil-
Spleen 2.30 0.70
laries and sinusoidal veins in the region adjacent to the
Ovaries 0.19 0.06
Testes 0.04 0.02 growth plate and there is also a paucity of phagocytes
Total body 0.19 0.05 ( Figs. 12-23 and 12-24). The osteomyelitis is often sec-
ondary to staphylococcal skin or mucosal infection.

Figure 12-22 Acute appendicitis: Tc-99m fanolesomab (NeutroSpec). Patient with clinically
uncertain but suspected acute appendicitis. The scan demonstrates focal activity in the right
lower quadrant of the abdomen at 30 minutes (arrow), which increases in intensity at one hour.
(Used with permission. Love C, Palestro CR: Radionuclide imaging of infection. J Nucl Med Technol
32:47–57, 2004.)
 Infection and Inflammation 405

In the adult, osteomyelitis from hematogenous spread

Box 12-9 Gallium-67: Clinical Indications rarely involves the long bones because red marrow has
been replaced by adipose tissue (yellow marrow). It
Severe leukopenia (<3000/m3 leukocytes) often occurs in vertebral bodies, where the marrow is
Sarcoidosis cellular with an abundant vascular supply. Infection
Idiopathic pulmonary fibrosis begins near the anterior longitudinal ligament and
Pulmonary drug reactions (amiodarone, bleomycin) spreads to adjacent vertebrae by direct extension
Pneumocystis carinii through the disk space or via communicating venous
Suspected low grade chronic infections channels. Because the adult disk does not have a vascu-
Immunosuppressed patients with pulmonary infection
lar supply, disk space infection is always due to
Fever of unknown etiology if both infection and tumor
are in the differential diagnosis
osteomyelitis in an adjacent vertebra. The initiating
Malignant external otitis event is usually septicemia from urinary tract infection,
bacterial endocarditis, or intravenous drug abuse.

Extension from a Contiguous Site of Infection

The most common cause of osteomyelitis is direct exten-
Table 12-7 Specific Indications for In-111 Oxine or sion from overlying soft tissue infection, which is usually
Tc-99m HMPAO Leukocytes
secondary to trauma, radiation therapy, burns, or pressure
sores. In patients with diabetes and vascular insufficiency,
In-111 oxine leukocytes Tc-99m HMPAO leukocytes organisms enter the soft tissues through a cutaneous ulcer,
often in the foot, causing cellulitis and then osteomyelitis.
Intra-abdominal infection Pediatric patients
Cardiovascular infection Inflammatory bowel disease
Diabetic mid-foot and Osteomyelitis of extremities Direct Introduction of Organisms into Bone
hind-foot osteomyelitis* including the feet in non-diabetic Direct inoculation of infection may occur with open frac-
Inflammatary bowel disease patients tures, open surgical reduction of closed fractures, or pen-
Hip and knee prosthesis* Diabetic forefoot osteomyelitis
etrating trauma by foreign bodies. Osteomyelitis may
Orthopedic hardware,
prior fracture, infection* also arise from perioperative contamination of bone dur-
ing surgery for nontraumatic orthopedic disorders, as in
*In conjunction with Tc-99m SC bone marrow scan. laminectomy, diskectomy, or placement of a joint pros-
thesis.

Figure 12-23 Vascular supply of long bones. In the infant and until 18 months of age, small
vessels perforate the physis to enter the epiphysis. After 18 months and during childhood, the
perforating vessels involute. The epiphysis and metaphysis then have separate blood supplies.
Following closure of the physis, branches of the nutrient artery extend to the end of bone (adult)
and the principal blood supply to the end of long bones is again from the nutrient artery in the
medullary canal. The periosteal artery supplies the outer cortex, whereas branches of the nutrient
artery supply the inner cortex.
406 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-24 Pathophysiology of hematogenous osteomyelitis (child). A, Bacterial embolization

occurs via the nutrient artery, and bacteria lodge in the terminal blood supply in the metaphysis.
B, After the infection is established, it expands within the medullary canal toward the cortex and
diaphysis. The physis serves as an effective barrier. C, The infection then extends through the
vascular channels to the cortex to elevate and strip the periosteum from the cortex and periosteal
new bone forms. The bond between the periosteum and perichondrium at the physis prevents
extension of the infection into the joint.

hands and feet. Radiographic methods and nuclear med-

Table 12-8 Diagnosis of Osteomyelitis: Accuracy icine scans have become important for confirming or
of Different Scintigraphic and MRI
excluding the diagnosis ( Table 12-8).
Methods
Conventional Imaging
Sensitivity Specificity Plain film radiography should be performed whenever
Type of study (%) (%)
osteomyelitis is suspected because it can be diagnostic.
Three-phase bone scan 94 95 However, the characteristic changes of permeative radio-
(normal x-ray) lucencies, destructive changes, and periosteal new bone
Three-phase bone scan 95 33 formation may take 10–14 days to develop and are not
(underlying bone disease)
always specific.
Gallium-67 81 69
Indium-111 oxine leukocytes 88 85 MRI can detect marrow changes of osteomyelitis (i.e.,
Technetium-99m HMPAO 87 81 low-signal intensity on T1-weighted images and high
Leukocytes (vertebral) 40 90 signal intensity on T2-weighted images and enhance-
Leukocytes + bone marrow 95 90 ment with gadolinium). Secondary changes such as
Antigranulocyte 92 88
sinus tracts and cortical interruption increase the cer-
Magnetic resonance imaging 95 87
tainty. The sensitivity and specificity is reported to be
high. However, any disease that replaces bone marrow
and results in increased tissue water (e.g., healing frac-
Osteomyelitis initially acquired as a child or in adult- tures, tumors, and Charcot joints) may not be distin-
hood may be manifested later as intermittent or per- guishable from infection. Artifacts caused by joint
sistent drainage from sinus tracts communicating with implants can degrade images sufficiently to make diagno-
the involved bone, usually the femur, tibia, or humerus, sis impossible.
or as a soft tissue infection overlying it. Signs of infec-
tion may recur after years of quiescence. Scintigraphy
The best scintigraphic method for diagnosis of osteo-
Clinical Diagnosis myelitis depends to some extent on the particular clini-
Biopsy with culture is the most definitive test for the cal situation ( see Box 12-9).
diagnosis of osteomyelitis, but this is invasive and often Bone Scan
contraindicated. Noninfected bone may become con- In patients who do not have an associated underlying
taminated by overlying soft tissue infection and there is bone condition, a three-phase bone scan should be per-
risk of pathological fracture in the small bones of the formed first. Its overall sensitivity for the diagnosis

Figure 12-25 Combined Ga-67 and bone scan: confirm or
exclude osteomyelitis of the spine. Fever after laminectomy raised
the question of infection. Vertebral Ga-67 uptake (bottom) was
judged to be less than that seen on the Tc-99m medronate (MDP)
bone scan (top), and the study was interpreted as negative for
vertebral osteomyelitis. The low-grade Ga-67 uptake was likely the
result of reactive healing bone.
of osteomyelitis is greater than 95%. A negative study
excludes osteomyelitis with a high degree of certainty.
The one exception is in neonates who on occasion may
have false-negative bone scintigraphy. However, the
specificity of the bone scan is poor in patients with under-
lying bone disease such as fractures, orthopedic implants,
and neuropathic joints (30–50%) (see Table 12-8).
Gallium-67 Citrate
Because Ga-67 is taken up by normal bone as well as mar-
row, locally increased Ga-67 uptake occurs at sites of
increased bone turnover, similar to that seen with a bone
scan. Thus, false-positive interpretations may result in
patients with underlying bone conditions. The speci-
ficity of the Ga-67 scan can be increased if interpreted in
conjunction with a bone scan.
The accepted criteria for diagnosis of osteomyelitis
using Ga-67 scintigraphy in conjunction with a bone
scan are: (1) if Ga-67 uptake is greater than that seen on
Infection and Inflammation 407
Figure 12-26 Osteomyelitis of diabetic foot: In-111 leukocytes
and bone scan. Diabetes, peripheral vascular disease, cellulitis, and
infection in the region of the first metatarsal. Left, Two-hour
delayed bone scan shows marked increased uptake in the distal
first metatarsal. The first two phases (not shown) were also
positive. Right, In-111 oxine leukocyte study shows intense
uptake in the same distal metatarsal. No uptake is noted in other
areas of the foot that had increased uptake on the bone scan (i.e.,
the distal phalanx of the first toe and the second distal metatarsal).
the bone scan or (2) if the Ga-67 and bone scan distribu-
tion are incongruent in distribution (Fig. 12-25). If Ga-67
uptake is less than seen on the bone scan, it is negative
for osteomyelitis. A similar degree of uptake on both
studies is considered equivocal for the diagnosis.
However, the accuracy of the combined two studies is
still inferior to that of labeled leukocytes (see Table 12-8).
Radiolabeled Leukocytes
The reported accuracy of radiolabeled leukocytes
for diagnosis of osteomyelitis has varied considerably.
Attempts have been made to improve the accuracy by
interpreting the study in conjunction with a bone scan,
similar to the approach used with Ga-67 imaging ( Figs.
12-16, 12-26, and 12-27). However, this has not appre-
ciably improved the overall accuracy, particularly in
the problematic diabetic foot, spine, and hip and knee
prostheses.
Bone Marrow Scan in Conjunction with Leuko-
cyte Scan An underlying assumption of leukocyte scan
interpretation for osteomyelitis is that bone marrow
distribution is uniform and symmetrical and that an area
of focally increased uptake is diagnostic of infection.
However, when marrow distribution is altered, there is
potential for misinterpreting focal uptake as infection.
A solution to the problem of altered marrow distribu-
tion is to interpret the leukocyte scan in conjunction with
a bone marrow scan. The Tc-99m SC marrow scan can
serve as a template for the distribution of marrow in that
patient. With infection, the radiolabeled leukocyte study
will be discordant with the marrow scan ( i.e., focal
increased uptake on the leukocyte study), whereas the
marrow scan will show decreased or normal uptake. This
408 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-27 Metatarsal osteomyelitis: positive three-phase bone scan, negative In-111 leukocytes.
A, Flow study shows increased flow in the region of the distal left mid-foot. B, Left, A 3-hour delayed
image shows increased uptake by the third metatarsal. Diffuse ankle uptake is also noted. Right,
The In-111-leukocyte study is negative for infection. Radiograph showed a metatarsal fracture.

spatial incongruity is diagnostic of infection. With no Confirmation or exclusion of the diagnosis of

infection, the leukocyte and marrow scan will have similar
distribution.
Diabetic Foot
Insensitivity of the neuropathic foot to pain often results
in asymptomatic trauma, fractures, ulcers, infection, and
delay in diagnosis. Foot ulcers serve as a common portal
of entry for infection. The diagnosis of osteomyelitis in
the setting of soft tissue infection is a common but diffi-
cult diagnostic problem in the diabetic. Radiographs
and MRI are often nonspecific for infection.
The three-phase bone scan is frequently not helpful
because it is so often abnormal due to overlying soft tissue
infection (increased flow) (see Fig. 12-27), fractures, neu-
ropathic Charcot’s joints (can be three-phase positive),
and degenerative changes. Although the sensitivity of the
bone scan for diagnosis of osteomyelitis approaches
100%, its specificity is much poorer ( see Table 12-8).
Delayed imaging at 24 hours (fourth phase) is reported to
improve accuracy, but false positives are still a problem.
osteomyelitis often requires leukocyte scintigraphy.
However, differentiating bone infection from overlying
soft tissue infection can be problematic with In-111
leukocyte because of the paucity of anatomical informa-
tion on the leukocyte scan ( see Fig. 12-26). The
improved resolution of Tc-99m HMPAO leukocytes with
low-level distribution in soft tissue is an advantage over
In-111 leukocytes. Tc-99m HMPAO is particularly useful
in the distal foot (Figs. 12-28 and 12-29).
The foot proximal to the metatarsals is more problem-
atic diagnostically. Although red marrow is not present
in the distal extremities in adults, neuropathic Charcot
joints form marrow and accumulate leukocytes.
Fractures may also stimulate marrow formation. Thus,
marrow scintigraphy in conjunction with In-111 leuko-
cyte imaging is important for evaluating the mid- or hind-
foot of the diabetic. One approach is to acquire the
leukocyte and bone marrow studies on separate days. In
that case Tc-99m HMPAO would be preferable. However,
an alternative approach and perhaps preferable is to do
simultaneous dual-isotope leukocyte scintigraphy and
 Infection and Inflammation 409

bone marrow scan using In-111 oxine labeled leukocytes
and Tc-99m SC for the marrow scans (Fig. 12-30).
Vertebral Osteomyelitis
The most common route of vertebral infection is hematoge-
nous spread via the arterial or venous system. However,
postoperative infections are not rare with direct implanta-
tion of microorganisms into the intervertebral disc.
Microorganisms lodge at different sites depending on
patient age. Below 4 years of age, end arteries perforate
the vertebral body end plates and enter the disc space pro-
ducing discitis. In adults, the richest network of nutrient
arterioles is localized in the subchondral region of the

Figure 12-28 Osteomyelitis of distal phalanx: Tc-99m HMPAO leukocytes and bone scan.
Diabetic with purulent drainage of the distal second digit of the right foot. A, Bone scan shows
increased uptake on the distal second digit of the right. The first two phases were also positive.
B, Tc-99m HMPAO leukocytes were positive in the distal second digit, consistent with osteomyelitis.
Other uptake on the bone scan was negative on the leukocyte study.

Figure 12-29 Tc-99m HMPAO soft tissue infection. Clinically ulcers of the right lateral malleolus
and heal of right foot. Soft tissue infection was diagnosed, not osteomyelitis, was diagnosed with the
leukocyte study. The low-grade soft tissue of Tc-99m HMPAO distribution makes it easier to
differentiate osteomyelitis from soft tissue infection than with In-111 leukocytes.
410 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-30 Combined In-111 leukocyte and Tc-99m SC studies for possible osteomyelitis of the
first phalanx. Recent bunionectomy and internal fixation. The joint became infected, was treated with
antibiotics, but C-reactive protein was persistently elevated and concern for osteomyelitis. Plantar
images of feet. Tc-99m SC (left) and In-111 oxine leukocytes (right). Both have similar extended
pattern of uptake (concordant) in region of first MP on the left. Negative for bone infection.

vertebral body, similar to the vascular tree in the childhood

metaphysis. This is the most common site of infection in
the adult. The infection is primarily spondylitis with sec-
ondary spread into the disc space. The infection spreads
from the anterior subchondral focus through the vertebral
end plate into the intervertebral disc. Later it destroys the
neighboring end plate, involves the opposite vertebral
body, and finally may extend to adjacent soft tissue and
result in epidural or paravertebral abscess formation.

Bone Scan
The typical pattern of discitis is increased blood flow,
blood pool, and delayed uptake involving adjacent con-
tiguous ends of adjoining vertebral bodies. With isolated
osteomyelitis, the bone scan will be three-phase positive.
However, blood flow and blood pool images can be diffi-
cult to evaluate in the thoracic spine because of normal
cardiac, pulmonary, and vascular structures.

Figure 12-31 Vertebral osteomyelitis: false negative. A cold
defect at L5 on indium-111 leukocyte scintigraphy. Biopsy was
necessary to make the diagnosis of osteomyelitis.
Leukocyte Scintigraphy
Multiple studies have now reported a high false negative
rate for both In-111 oxine and Tc-99m HMPAO radio-
labeled leukocytes in the spine. False-negative results
occur in approximately 40% of studies. The labeled
leukocyte study may be normal or there may be a pho-
topenic or cold defect at the site ( Fig. 12-31). Thus,
infection cannot be differentiated from metastasis, frac-

Figure 12-32 Disk space infection: Ga-67. Clinically suspected
disk infection. A, Magnetic resonance imaging showed only a
narrowed interspace with evidence of degenerative disc disease.
B, Posterior view, prominent Ga-67 uptake in the region of the L3-
4 disc space.
ture, Paget disease, surgical defects, or irradiation. The
explanation for this finding is uncertain, but may be due
to bone marrow thrombosis and infarction.
Ga-67 Scintigraphy
Ga-67 is probably superior to leukocyte scintigraphy for
vertebral osteomyelitis. However, false positives may
Infection and Inflammation 411
result due to reactive bone. To minimize this problem,
Ga-67 should be done in conjunction with a bone scan
using the criteria discussed previously ( see Fig 12-25).
An associated paraspinal abscess or disc infection can
be also diagnosed with Ga-67 scintigraphy ( Fig. 12-32).
Other Scintigraphy
F-18 FDG may have a role in this setting, although reim-
bursement is an issue at present. Early data suggest that
Tc-99m ciprofloxacin ( Infecton), still investigational,
may be useful for this indication.
Infected Joint Prostheses
The infection rate after primary hip or knee replacement
is only 1% and after revision surgery is less than 3%.
However, when present, an infected prosthesis pro-
duces severe morbidity. Diagnosis of an infected pros-
thetic joint can be difficult because the symptoms and
signs are often indolent. Radiographs also have poor
sensitivity for early detection. Joint aspiration also has
a low sensitivity for diagnosis.
Bone Scan
Characteristic bone scan findings have been described to
differentiate loosening and infection of hip prostheses.
In patients studied for more than 12 months after inser-
tion of cemented total hip prostheses, diffuse uptake sur-
rounding the femoral component suggests infection.
However, this pattern is only moderately predictive.
Cementless or porous coated prosthesis depend on bony
ingrowth for stabilization. Ongoing new bone formation
is part of the fixation process, causing periprosthetic
uptake on bone scintigraphy in a variable pattern for
a prolonged period, making interpretation more difficult.
Knee prostheses are a particular problem for bone
scintigraphy. More than half of all femoral components
and three-fourths of all tibial components show peripros-
thetic uptake more than 12 months after placement.
Thus, for patients with cementless hip replacement or
total knee replacement, bone scintigraphy is most useful
when the scan is normal or when serial studies over time
are available for comparison.
Ga-67 Scintigraphy
Even in conjunction with bone scintigraphy Ga-67 is
only moderately accurate, reportedly 80%, in the diagno-
sis of infected joint prostheses.
Leukocyte Scintigraphy
The reported accuracy of leukocyte scintigraphy in combi-
nation of the bone scan imaging has been variable.
Evidence strongly suggests that superior diagnostic accu-
racy results when the leukocyte scan is interpreted in com-
bination with bone marrow scintigraphy. The Tc-99m
412 NUCLEAR MEDICINE: THE REQUISITES

Figure 12-33 Infected hip prosthesis: In-111 leukocyte and Tc-99m SC marrow study. Left, Tc-99m
medronate (MDP) bone scan shows increased uptake in the region of the right hip prosthesis laterally
consistent with heterotopic calcification. Middle, Indium-111 leukocyte study shows focal intense
uptake just lateral to the femoral head and more diffuse uptake within the joint space consistent with
infection. Right, Tc-99m SC marrow study shows a normal bone marrow distribution with cold head
of the femur consistent with prosthesis. The discordance of the bone marrow and In-111 leukocyte
study indicates an infected prosthesis.

bone marrow scan serves as a template for the patient’s
normal marrow distribution. The normal distribution of
marrow in the elderly population is variable. Placement of
an orthopedic device inevitably results in marrow displace-
ment in an unpredictable manner. The marrow study
avoids the potential for a false-positive study, that is, inter-
pretation of focal uptake as infection when it is merely dis-
placed marrow distribution.
If performed on different days, Tc-99m HMPAO leuko-
cytes can be used in conjunction with Tc-99m SC.
Simultaneous dual-isotope acquisition in the identical pro-
jections is advantageous and can be done with the dual-
isotope scanning In-111 oxine leukocytes and the Tc-99m
SC marrow scan. Discordance of uptake on the leukocyte
study and the marrow scintigraphy is diagnostic ( Fig.
12-33). Accuracy is reported to be greater than 90%.
Response to Therapy
Both Ga-67 and leukocyte scintigraphy can also be useful
for monitoring response to therapy (e.g., to ascertain
that infection has been controlled prior to surgical
replacement of a new prosthesis). Scintigraphic findings
of infection should revert to normal by 2–8 weeks of
appropriate antibiotic therapy.
Intra-Abdominal Infection
The diagnosis of intra-abdominal infection is often made by
CT with directed interventional aspiration. However, in
patients with nonlocalizing symptoms and negative con-
ventional imaging, scintigraphy has clinical value. Gallium-
67 has been used to diagnose intra-abdominal infection;
however, it has numerous drawbacks, particularly normal
bowel clearance and delayed imaging. In-111–oxine
leukocytes have a distinct advantage (see Tables 12-5 and
12-7). The radiopharmaceutical is not cleared in the
abdomen. Thus any intra-abdominal localization is likely
due to infection. Combined data from three large series
reported an overall sensitivity of approximately 90% for
detection and localization of intra-abdominal infection.
Early imaging at 4 hours with In-111 leukocytes has
a reduced sensitivity for the detection of infection com-
pared to 24 hours; however, in some situations, early
imaging may expedite making the diagnosis (e.g., in
acutely ill patients with suspected acute appendicitis,
diverticulitis, and ischemic bowel disease). In these
acute illnesses, early intensive leukocyte uptake is likely
due to the increased blood flow to the site of infection
and the marked inflammatory response with leukocyte
infiltration.
Tc-99m HMPAO label has the advantages of its supe-
rior image quality and preferential labeling of granulo-
cytes, resulting in rapid uptake in pyogenic infections.
However, its hepatobiliary and renal clearance requires
imaging at 1–2 hours. A large study found that the sen-
sitivity for detecting abdominal infection and inflamma-
tory disease was 88% at 30 minutes and 95% at 2 hours.
Delayed imaging can sometimes be helpful to confirm
that the early detected abnormal activity is a fixed pat-
tern. A shifting pattern of activity over time implies
intraluminal transit of labeled leukocytes, as seen with
inflammatory or ischemic bowel disease, fistula,
abscess in communication with bowel, or false-positive
causes (e.g., swallowed leukocytes from sinus or tra-
cheobronchial infection) ( see Box 12-7). Abnormal
leukocyte uptake has been described in a variety of
 Infection and Inflammation 413

Table 12-9 Diagnostic Accuracy of Tc-99m Inflammatory Bowel Disease

Fanolesomab (NeutroSpec) for Acute
Appendicitis Both ulcerative colitis and Crohn’s disease (granuloma-
tous or regional enteritis) are characterized by intestinal
inflammation that can be detected with leukocyte
Percentages (95% CI)
scintigraphy ( Fig. 12-34). Scintigraphy has been used to
200-patient make the diagnosis, to determine disease distribution,
multicenter Study and to confirm relapse. Scintigraphy can aid in the eval-
trial Blinded readers investigators
uation of regions hard to see with endoscopy and can
Sensitivity 75 (62–86) 91 (80–97) also monitor the effectiveness of therapy. Studies have
Specificity 93 (87–97) 86 (79–91) shown good correlation between the site and amount of
Accuracy 87 (82–92) 87 (81–91) leukocyte uptake compared with the endoscopic and
Positive predictive value 82 (69–91) 74 (62–84)
radiological localization.
Negative predictive value 90 (84–94) 96 (90–99)
Leukocyte scintigraphy can differentiate reactivation
of inflammatory bowel disease from abscess formation
resulting from bowel perforation. The latter is a serious
clinical problem, requiring very different therapy (surgi-
noninfectious inflammatory diseases (e.g., pancreatitis, cal rather than medical). Radiolabeled leukocyte uptake
acute cholecystitis, polyarteritis nodosa, and rheuma- in an abscess is typically focal, whereas uptake in
toid vasculitis). inflamed bowel typically follows the contour of the intes-
Leukocyte uptake occurs in ischemic colitis (common tinal wall. In-111 leukocyte imaging can also aid the
in elderly patients), pseudomembranous colitis (anti- surgeon in determining if a partially obstructed bowel is
biotic related), and bowel infarction. Inactive colitis is the result of active inflammation or is a fibrotic stricture.
negative on scintigraphy. In the latter case, surgery is indicated.
Tc-99m fanolesomab ( NeutroSpec) has recently been In-111 oxine leukocyte imaging should be performed
approved for the diagnosis of acute appendicitis when at 4 hours rather than the usual imaging time of 24 hours
the clinical diagnosis is uncertain. A multicenter trial of because of shedding of the inflamed leukocytes into the
200 patients with equivocal signs and symptoms of bowel lumen from the inflammatory sites and subsequent
appendicitis were investigated. The overall accuracy peristalsis may result in incorrect assignment of disease to
was good with a positive predictive value of 82% and sites distal to the true lesion. Tc-99m HMPAO imaging
negative predictive value of 90% ( Table 12-9). Some of must be performed at 1–2 hours prior to intra-abdominal
the false positives may have been other sites of infection. hepatobiliary and genitourinary radiotracer clearance.
The high negative predictive value saves patients from Comparative studies have indicated that Tc-99m
admission or surgery. HMPAO-labeled leukocytes are advantageous over In-111–
oxine leukocytes to diagnose inflammatory bowel disease
because of its superior image resolution enabling disease
localization to specific bowel segments and better identifi-
cation of small bowel disease. Accurate results with
Tc-99m HMPAO leukocytes are possible by 1 hour after
injection. Crohn disease and ulcerative colitis can usually
be distinguished from each other by the distribution of dis-
ease activity. Rectal sparing, small bowel involvement,
and skip areas suggest Crohn disease, whereas continuous
colonic involvement from the rectum without small
bowel involvement suggests ulcerative colitis.

Figure 12-34 Crohn disease: 4-hour In-111 leukocytes. Several
year history of regional ileitis. Two months of recurrent and
worsening symptoms. Scintigraphy confirms active inflammation
of ileum.
Renal Disease
Ga-67 has been used to diagnose diffuse interstitial
nephritis and localized infection, but delayed imaging
must be performed because of its clearance through the
urinary tract during the initial 24 hours after injection.
Renal parenchymal infection such as pyelonephritis or
diffuse interstitial nephritis ( Fig. 12-35), lobar nephronia
414 NUCLEAR MEDICINE: THE REQUISITES
(focal interstitial nephritis), and perirenal infections
( Fig. 12-36) have been diagnosed with Ga-67. However,
with renal or hepatic failure and iron overload, renal
uptake is increased without infection.
For the most part Ga-67 has been superceded by leuko-
cyte imaging Tc-99m DMSA and, importantly, biopsy.
Radiolabeled leukocytes also accumulate at sites of acute
pyelonephritis, focal nephritis (lobar nephronia), and renal
Figure 12-35 Interstitial nephritis: Ga-67. Bilateral intense
renal gallium-67 uptake is seen at 48 hours (posterior view).
Normal uptake is seen in the liver, bone, and bone marrow.
Figure 12-36 Perirenal abscess: Ga-67. Fever and pain
developed after renal stone removal and nephrostomy. Focal
increased Ga-67 uptake is seen just inferior to the spleen and
adjacent to the left kidney, consistent with an abscess.
or perirenal abscess. However, In-111–labeled leukocytes
have limited utility for evaluation of renal transplants
because all exhibit uptake, regardless of the presence or
absence of clinically significant disease or rejection.
Cardiovascular Disease
In-111–labeled leukocytes are reported not to be highly
sensitive for detecting subacute bacterial endocarditis.
The vegetative lesions contain high concentrations of
bacteria, platelets, and fibrin adherent to damaged valvu-
lar endothelium, but relatively few leukocytes. However,
it occasionally can be useful and SPECT may be helpful.
Leukocyte uptake occurs in acute myocardial infarction
and cardiac transplant rejection, but leukocyte scintigra-
phy is used for making these diagnoses.
Infection of arterial prosthetic grafts (e.g., femoro-
popliteal and aortofemoral) is associated with significant
morbidity and mortality. Ultrasound, CT, and MRI are often
unable to distinguish infection from aseptic fluid collections
around the graft. Prompt diagnosis of graft infection is criti-
cal but often delayed because of their indolent and insidious
course. Radiolabeled leukocytes can detect surgical pros-
thetic graft infection (Fig. 12-14). In-111 has a theoretical
advantage of having no blood pool distribution, a potential
problem with Tc-99m HMPAO. However, early imaging
with Tc-99m HMPAO at 5 minutes, 30 minutes, and 3 hours
has been shown to be quite accurate for confirming the
diagnosis. Infected grafts have persistent uptake.
Pulmonary Infections
Pulmonary uptake of radiolabeled leukocytes should be
interpreted cautiously. Low-grade diffuse uptake has
been associated with a variety of noninfectious causes,
including atelectasis, congestive heart failure, and adult
respiratory distress syndrome, and therefore should not
be considered diagnostic of infection.
Focal intense uptake is likely to be due to infection
( see Fig. 12-16). Tuberculosis and fungal infections may
be detected by In-111 leukocytes, but with a generally
lower sensitivity than Ga-67.
Ga-67 citrate accumulates in virtually all types of pul-
monary infections and inflammatory diseases, including
pneumonia, lung abscess, and tuberculosis. However,
the usual clinical role for Ga-67 is for the most part in
more chronic diseases, such as sarcoidosis (see Fig. 12-4),
idiopathic pulmonary fibrosis, Pneumocystis carinii (see
Fig. 12-37), and for therapeutic drug-induced pulmonary
disease (Boxes 12-10 to 12-12).
Sarcoidosis
Sarcoidosis a chronic granulomatous multisystem disease
of unknown etiology characterized by an accumulation
of T-lymphocytes, mononuclear phagocytes, and non-
 Infection and Inflammation 415

Box 12-12 Therapeutic Agents Associated

with Gallium-67 Lung Uptake

Bleomycin
Amiodarone
Busulfan
Nitrofurantoin
Cyclophosphamide
Methotrexate
Nitrosourea

caseating epithelioid granulomas. The lungs are the

organ most commonly involved. Manifestations include
hilar and mediastinal adenopathy, endobronchial granu-
loma formation, interstitial or alveolar pulmonary infil-
trates, or pulmonary fibrosis.
Systemic symptoms such as weight loss, fatigue, weak-
ness, malaise, and fever are common ( 40%). Pulmonary
disease predominates. However, disease can involve any
Figure 12-37 Pneumocystis carinii-Ga-67. Diffuse intense
homogeneous, as seen in this study, or diffuse heterogeneous uptake
organ of the body, most commonly the liver and spleen.
is typical of this infection. The characteristic pattern is different from Extrapulmonary manifestations of the skin, eyes, heart,
that of most other common pulmonary infections in AIDS patients. central nervous system, bones, and muscle are not rare.
Central nervous system and cardiac involvement when
present may lead to death.
Box 12-10 Scintigraphic Diagnosis The initial presentation is usually pulmonary and one-
of Osteomyelitis in Different half of patients present with complaints of dyspnea and
Clinical Situations dry cough. However, as many as 20% are asymptomatic
and have only an abnormal chest radiograph at initial
Normal x-ray: three-phase bone scan detection. The clinical course is variable. Spontaneous
Neonates: three-phase bone scan; if negative, Tc-99m resolution occurs in about one-third of patients. Another
HMPAO 30–40% have a smoldering or progressively worsening
Suspected osteomyelitis in non-marrow-containing course, 20% develop permanent loss of lung function,
skeleton (distal extremities): bone scan + leukocyte and 5–10% die of respiratory failure.
study There are four types of x-ray findings ( Box 12-13): no
Suspected osteomyelitis in bone marrow-containing abnormality ( type 0), bilateral hilar adenopathy ( type I),
skeleton (hips and knees): marrow scan + leukocyte bilateral adenopathy with diffuse parenchymal abnor-
study
malities (type II), and diffuse parenchymal changes with-
Suspected vertebral osteomyelitis: gallium-67
out hilar adenopathy (type III ). The alveolitis of
sarcoidosis is manifested on radiographs as an infiltrative
process. Although patients with type I radiographs tend
Box 12-11 Ga-67 Uptake in Interstitial to have a reversible form of the disease and those with
and Granulomatous types II and III usually have chronic progressive disease,
Pulmonary Diseases these patterns are not necessarily consecutive stages.
Diagnosis is based on a combination of clinical, radio-
graphic, and histological findings. The chest x-ray,
Tuberculosis
although characteristic, is not diagnostic because the typi-
Histoplasmosis
Sarcoidosis
cal bilateral hilar adenopathy may be seen with other dis-
Idiopathic pulmonary fibrosis eases. Biopsy evidence of a mononuclear cell granulo-
Pneumocystis carinii matous inflammatory process is mandatory for definitive
Cytomegalovirus diagnosis. Because the lung is so frequently involved, it is
Pneumonoconioses (asbestosis, silicosis) the most common biopsy site, usually via fiberoptic bron-
Hypersensitivity pneumonitis choscopy. Biopsy may be performed on any involved
organ.
416 NUCLEAR MEDICINE: THE REQUISITES
Box 12-13 Classification of Chest
Radiographic Findings in
Sarcoidosis
Type Radiographic Findings
0 Hilar and/or mediastinal node enlargement
with normal lung parenchyma
II Hilar and/or mediastinal node enlargement
and diffuse interstitial pulmonary disease
III Diffuse pulmonary disease without node
involvement
IV Pulmonary fibrosis
Other tests and methods are used to diagnose sar-
coidosis. Bronchoalveolar lavage is an accurate
method of making the diagnosis. The finding of an
increase in the relative and absolute numbers of T-lym-
phocytes, monocytes, and macrophages in the lung is
used as an indication for therapy. The serum
angiotensin-converting enzyme is negative in two-thirds
of patients and false-positive results are common. The
Kveim-Siltzbach test requires intradermal injection of
human sarcoid tissue. A nodule develops at the injection
site in 4–6 weeks in patients with sarcoidosis and biopsy
reveals noncaseating granulomas in 70–80%. This is not
commonly used today.
Many patients require no specific therapy. However,
those with severe active disease are often treated with
glucocorticoids that suppress the activated T-cells at the
disease site and the clinical manifestations of the disease.
Steroids are only administered if the disease is active.
The chest radiograph is not a sensitive indicator of dis-
ease activity. Bronchoalveolar lavage and Ga-67 scans
are used as indicators of disease activity.
Gallium-67 Scintigraphy
Ga-67 scans are positive in most patients with active sar-
coidosis. Scintigraphy has been used to assess the mag-
nitude of alveolitis, to guide lung biopsy, and to choose
the pulmonary segments for bronchoalveolar lavage. It
can distinguish active granuloma formation and alveolitis
from inactive disease and fibrotic changes. Increased
Ga-67 uptake in the lungs is more than 90% sensitive for
clinically active disease. Scans are negative in inactive
cases.
Ga-67 is more sensitive than a chest radiograph for
detecting early disease. Pulmonary uptake on scintigra-
phy can be seen before characteristic abnormalities are
present on radiographs. One third of patients have nor-
mal radiographs at this stage of disease. Patients with
a normal Ga-67 scan nearly always have a negative
biopsy. Patients with a diagnosis of sarcoidosis and an
abnormal chest radiograph, but inactive disease, have
a negative Ga-67 scan. In these cases the abnormal radio-
graph represents past, not present, disease.
There is generally good correlation between Ga-67
lung uptake and lymphocyte content in lavage fluid. The
Ga-67 scan’s primary use has been to evaluate for disease
activity and to guide therapeutic decisions. It is a sensitive
indicator of treatment response, superior to clinical symp-
toms, chest radiograph, and pulmonary function tests.
Scintigraphic Patterns
In early disease the Ga-67 scan often shows bilateral hilar
and paratracheal uptake ( lambda sign) ( see Fig. 12-4B).
Pulmonary parenchymal uptake when present is intense
and symmetrical ( see Fig. 12-4C) and may or may not be
associated with hilar and mediastinal involvement.
Patients with malignant lymphoma usually have asym-
metrical hilar or mediastinal uptake, often involving the
anterior mediastinal and paratracheal nodes. Although
paraaortic, mesenteric, and retroperitoneal lymph node
involvement may be seen in sarcoidosis, uptake is much
more common in lymphoma.
Prominent uptake in the nasopharyngeal region, the
parotid, salivary, and lacrimal glands is common ( panda
sign) ( see Fig. 12-4C). The combination of ocular
involvement (iritis or iridocyclitis) with accompanying
lacrimal gland inflammation and bilateral salivary gland
involvement is known as uveoparotid fever.
The degree of pulmonary uptake of Ga-67 is judged
relative to uptake in the liver, bone marrow, and soft tis-
sue. Lung uptake greater than liver is positive for sar-
coidosis. Uptake less than soft tissue is negative. The
posterior view is preferable for estimating uptake.
Oblique views or SPECT can be useful for discerning
mediastinal and hilar uptake or confirming pulmonary
uptake when in question.
Semiquantitative indexes of Ga-67 uptake have been
proposed. Although more objective quantification is
desirable, uptake by normal overlying soft tissue, breasts,
heart, sternum, ribs, scapulae, and spine and bone mar-
row complicate quantification. F-18 FDG also has promi-
nent uptake in active sarcoidosis.
Idiopathic Interstitial Pulmonary Fibrosis
The etiology of idiopathic interstitial pulmonary fibrosis
is unknown. Typically the disease follows a progression
through stages of alveolitis, with derangement of the
alveolar-capillary units, leading to end-stage fibrotic
disease. Ga-67 uptake is seen in approximately 70% of
patients. Ga-67 has been used to monitor the course
of disease and response to therapy. The amount of
uptake correlates with the degree of cellular infiltration,
but does not predict the results of steroid treatment.
Pulmonary Drug Reactions
Some of the therapeutic drugs known to cause lung injury
and result in Ga-67 uptake include cytoxan, nitrofurantoin,

bleomycin, and amiodarone ( see Box 12-11). Contrast
used for lymphangiography may cause a chemical-induced
alveolitis. Ga-67 uptake is an early indicator of drug-
induced lung injury, before the radiograph is abnormal.
Immunosuppressed Patients
The clinical presentation, physical findings, and radiologi-
cal abnormalities in immunosuppressed patients are often
obscured by an impaired inflammatory response caused
by the underlying disease or therapy. Furthermore, many
of the organisms causing infection (e.g., Pneumocystis
carinii, Cryptococcus, and cytomegalovirus [CMV])
produce a minimal inflammatory response even in
healthy hosts.
Immunosuppression is seen most commonly in
patients with AIDS and those receiving drugs for cancer
chemotherapy or organ transplantation. The diagnostic
sensitivity of sonography, CT, and MRI depends on the
presence of normal anatomical markers, which can be
disrupted by disease or previous surgical procedures.
Monitoring the effects of therapy is complicated by the
slowness of response, the lack of microbiological meth-
ods for assessing response, and the need for extended
courses of therapy for some opportunistic infections.
Ga-67 citrate can aid in the differential diagnosis of
pulmonary disorders in immunosuppressed patients
( Fig. 12-38). The patterns of Ga-67 pulmonary uptake
can be classified as diffuse parenchymal, focal parenchy-
mal, or nodal.
Diffuse Parenchymal Uptake
The first pulmonary manifestation of AIDS is often
Pneumocystis carinii pneumonia (PCP). The chest x-ray
Pulmonary Ga-67 uptake
Diffuse Focal Nodal
Intense Mild Bacterial MAI
pneumonia M. tuberculosis
Lymphoma
PCP CMV
MAI
Interstitial pneumonitis
Treated PCP
Severe PCP
Figure 12-38 Diagnostic decision tree: pulmonary Ga-67
uptake in the immunosuppressed patients. Ga, Gallium; MAI,
mycobacterium avium-intracellulare; PCP, pneumoocystis carinii
pneumonia; CMV, cytomegalovirus.
Infection and Inflammation 417
findings are usually abnormal, with bilateral diffuse infil-
trates originating from the hilum and extending periph-
erally. However, the radiograph may show a lobar or
nodular infiltrate or be normal.
Ga-67 scans are abnormal in approximately 90% of
cases. The scan is often positive before the chest radio-
graph becomes abnormal. The characteristic pattern in
PCP infection is diffuse bilateral pulmonary uptake ( see
Fig. 12-37), uniform or nonuniform, without nodal or
parotid uptake.
Increased Ga-67 uptake in an immunocompromised
patient may be from other causes, including CMV infec-
tion, bacterial pneumonitis, lymphocytic interstitial
pneumonitis, or the effects of various drug therapies.
The greater the Ga-67 uptake, the more likely the diagno-
sis is PCP. Uptake at initial presentation of PCP is
typically higher than that seen after the treatment of
recurrences. The prophylactic use of aerosolized pen-
tamidine therapy has resulted in atypical heterogeneous
patterns of uptake.
Cytomegalovirus Infection
The radiographic appearance in lymphoid interstitial
pneumonia may be normal or similar to that seen in PCP,
viral infections, or miliary tuberculosis. Ga-67 uptake is
only low-grade and diffuse without nodal uptake. This
may be accompanied by ocular uptake caused by retini-
tis, adrenal and renal uptake, persistent colon uptake
associated with diarrhea, and sometimes symmetrically
increased parotid uptake.
Focal Pulmonary Uptake
This is a less common pattern typically seen with bacte-
rial pneumonia. Corresponding infiltrates are seen on
chest radiographs. Intense Ga-67 uptake in a lobar con-
figuration in the absence of nodal and parotid uptake
suggests bacterial pneumonia. When multiple sites of
focal accumulation are present, aggressive infec-
tions caused by Actinomyces, Nocardia, and Aspergillus
should be considered.
Nodal uptake Mycobacterium avium-intracellulare
( MAI ) infection, tuberculosis, lymphoma, and occasion-
ally PCP will have nodal uptake in conjunction with
pulmonary uptake. Other causes are lymphadenitis,
cryptococcal infection, and herpes simplex. Infection
with MAI causes widespread disease in 25–50% of AIDS
patients and requires more aggressive therapy than that
used for tuberculosis. Patchy lung uptake with hilar and
nonhilar nodal ( axillary, inguinal) uptake suggests MAI.
Negative Ga-67 Uptake
The absence of Ga-67 uptake in conjunction with a nega-
tive chest radiograph excludes pulmonary infection with
a high degree of certainty. When the chest x-ray findings
are positive, particularly in a patient with deteriorating
418 NUCLEAR MEDICINE: THE REQUISITES
respiratory status, but Ga-67 is negative, Kaposi’s sar-
coma must be considered.
Intracerebral Infection
The differential diagnosis of an intracerebral mass in an
AIDS patient includes infection, most commonly toxo-
plasmosis and malignancy, most commonly lymphoma.
Generally Tl-201 has been used for this purpose.
Increased uptake greater than contralateral brain or scalp
is consistent with malignancy and low or no uptake is
characteristic of inflammatory disease.
Abdominal and Pelvic Infections
In-111-labeled leukocytes are superior to Ga-67 for diag-
nosing intestinal infection because of normal Ga-67 intes-
tinal clearance. Occasionally abnormal uptake will be
noted incidentally when pulmonary disease is being
assessed. In the immunosuppressed patient, proximal
small bowel uptake occurs with the protozoon infection
Cryptosporidium. When stool cultures are negative for
Salmonella or Shigella, diffuse colonic uptake that does
not change with time is probably the result of CMV infec-
tion or antibiotic-induced colitis. The findings of eye,
adrenal, esophageal, and low-grade pulmonary uptake
increases the certainty of CMV. Multifocal activity (e.g.,
paratracheal and bowel) is indicative of mycobacterial
infection.
Malignant External Otitis
This life-threatening infection occurs in diabetics sec-
ondary to Pseudomonas. Increased Ga-67 uptake can
differentiate this disease from other less serious causes
such as therapy-resistant external otitis. With malignant
otitis, increased uptake is seen in the temporal bone on
both bone scans and Ga-67 scans. The bone scan can
establish the initial diagnosis. Ga-67 scan is useful for
evaluating the effectiveness of therapy.
Fever of Unknown Origin
Fever of unknown origin is defined by clinicians as
a temperature of at least 38.3°C that occurs on more
than three occasions, remains without a diagnosed
cause for at least 3 weeks, and results in at least 7 days of
hospitalization. For patients who have not had recent
surgery, Ga-67 is a sensitive test for uncovering the
source of the fever. In addition to localizing acute infec-
tion, Ga-67 can detect chronic and indolent infections,
granulomatous infections, and even tumor sources of
fever. However, postoperative patients with fever are
better served with In-111-labeled leukocytes because
fever is most commonly caused by an acute infection.
Additionally In-111 leukocytes do not have the bowel
clearance problem of Ga-67 to confound intra-abdomi-
nal interpretation.
SUGGESTED READING
Blockman D, Knockaert D, Maes A, et al: Clinical value of 18F
fluoro-deoxyglucose positron emission tomography for patients
with fever of unknown origin. Clin Infect Dis 32:191–196, 2001.
Coleman RE, Datz FL: Detection of inflammatory disease using
radiolabeled cells. In Sandler M, Coleman RE, Wackers FJTh,
et al (eds): Diagnostic Nuclear Medicine, ed. 4, Baltimore,
Lippincott Williams & Wilkins, 2003.
Datz FL, Taylor AT Jr: Cell labeling: techniques and clinical util-
ity. In Freeman and Johnson’s Clinical Radionuclide Imaging,
ed. 3, Grune & Stratton, 1986.
McAfee JG, Samin A: In-111 labeled leukocytes: a review of prob-
lems in image interpretation. Radiology 155:221–229, 1985.
Meller J, Koster G, Liersch, et al: Chronic bacteria osteo-
myelitis: Prospective comparison of F-18 FDG imaging with
a dual-headed coincidence camera and In-111 label led autolo-
gous leucocyte scintigraphy. Eur J Nucl Med 29:53–60, 2001.
Merkel KD, Brown ML, Dewanjee MK, Fitzgerald RH Jr:
Comparison of indium-labeled-leukocyte imaging with sequen-
tial technetium-gallium scanning in the diagnosis of low-grade
musculoskeletal sepsis. J Bone Joint Surg 67A:465–476, 1985.
Oyen WJG, Boerman OC, van der Laken CJ, et al: The uptake
mechanism of inflammation- and infection-localizing agents.
Eur J Nucl Med 23:459–465, 1996.
Palestro CJ, Kim CK, Swyer AJ, et al: Total-hip arthroplasty:
Periprosthetic Indium-111-labeled leukocyte activity and com-
plementary Technetium-99m-sulfur colloid imaging in sus-
pected infection. J Nucl Med 31:1950–1955, 1990.
Palestro CJ, Kipper SL, Weiland F, et al: Osteomyelitis: diagnosis
with 99mTc-labeled antigranulocyte antibodies compared with
diagnosis with 111In-labeled leukocytes—Initial experience.
Radiology 223:758–764, 2002.
Palestro CJ, Mehta HH, Patel M, et al: Marrow versus infection
in the Charcot joint: indium-111 leukocyte and technetium-
99m sulfur colloid scintigraphy. J Nucl Med 39:346–350, 1998.
Palestro CJ, Thomas MB: Scintigraphic evaluation of the dia-
betic foot. In Leonard M. Freeman (ed). Nuc Med Annual
2000. Philadelphia, Lippincott Williams & Wilkins, 2000.
Peters AM, Danpure HJ, Osman S, et al: Clinical experience with
99mTc-hexamethylpropylene-amineoxime for labeling leuco-
cytes and imaging inflammation. Lancet 2525:946–948, 1986.
Rennen HJ, Boerman OC, Oyen WJ, et al: Specific and rapid
scintigraphic detection of infection with 99m Tc labeled inter-
leukin-8. J Nucl Med 42:117–123, 2001.
Rypins EB, Kipper SL, Weiland F, et al: 99mTc Anti-CD 15 mono-
clonal antibody ( LeuTech) Imaging improves diagnostic accu-
racy and clinical management in patients with appendicitis.
Ann Surg 235:232–239, 2002.
 13
CHAPTER Central Nervous
System

Cerebral Anatomy Nuclear medicine has imaged the central nervous system
Radiopharmaceuticals (CNS) for decades. Before the development of com-
Cerebral Blood Flow Radiopharmaceuticals puted tomography (CT ) in the 1970s, nuclear medicine
Tc-99m Cerebral Blood Flow Radiopharmaceuticals brain scans were the only noninvasive method for diag-
Glucose Metabolism—F-18 Fluorodeoxyglucose (FDG) nosing diseases of the brain including tumors, strokes,
Normal Distribution of Perfusion and Metabolism Agents and vascular anomalies. Although magnetic resonance
Methodology imaging ( MRI ) and CT are the most commonly utilized
Dosimetry brain-imaging modalities today, nuclear medicine offers
Clinical Applications of Cerebral Perfusion Imaging unique diagnostic information based on imaging physiol-
Dementia ogy. Single photon emission tomography ( SPECT ) and
Posterior Dementias positron emission tomography ( PET) can visualize physi-
Frontotemporal Dementia ologic changes of disease before anatomic changes can
Vascular Dementia be detected. Numerous radiopharmaceuticals have been
HIV Encephalopathy and AIDS-Dementia Complex used and some clinically important SPECT and PET
Stroke and Cerebrovascular Disease agents are listed in Box 13-1.
Brain Death
Epilepsy
Tumor Imaging Cerebral Anatomy
F-18 FDG PET Knowledge of brain anatomy is critical in understanding
SPECT Tumor Imaging patterns of disease and image interpretation. The brain
Movement Disorders consists of the two cerebral hemispheres above the ten-
Huntington’s Chorea torium and the cerebellum below in the posterior fossa.
Head Trauma The regions or lobes of the brain are illustrated in Figure
Psychiatric Diseases 13-1. The frontal lobe extends back to the central sulcus
Cisternography with the parietal lobe just posterior to it. The occipital
Radiopharmaceuticals lobe is most posterior, below the parieto-occipital sulcus.
Methods The temporal lobes are below the lateral fissure. Within
Pharmacokinetics these lobes, key functional centers or regions have been
Dosimetry identified, which are important when trying to assimilate
Clinical Applications clinical changes with anatomical and functional images
Hydrocephalus ( Fig. 13-2).
Cisternography Image Interpretation Studies such as dynamic radionuclide brain flow and
Surgical Shunt Patency brain death exam allow visualization of the vascular
Cerebrospinal Fluid Leak supply of the brain to a limited degree ( Figs. 13-3 and
13-4). Even more important for image interpretation is
familiarity with the cerebral regions these vessels sup-
ply ( Fig. 13-5).

419
420 NUCLEAR MEDICINE: THE REQUISITES

Motor cortex
Box 13-1 Brain SPECT and PET
Radiopharmaceuticals Used Sensory cortex
Clinically Wernicke’s
area

Radionuclide Broca’s
Peripheral Visual
area
Radiopharmaceutical Photopeak (keV) Fovea cortex

Blood–brain barrier
Tc-99m glucoheptonate 140
Primary
Tc-99m DTPA auditory
Brain perfusion cortex
I-123 iodoamphetamine (IMP) 159
Figure 13-2 Motor, sensory, visual, speech, and auditory
Tc-99m HMPAO 140 functional and associative centers of the brain.
Tc-99m methyl cysteinate dimer 140
(ECD)
Metabolism
F-18 fluorodeoxyglucose (FDG) 511 the 1970s, was Tc-99m pertechnetate. It did not normally
Brain tumor imaging cross the intact BBB. Accumulation in the choroid
Thallium-201 69–83, 167 (10%) plexus and salivary glands and slow blood pool clearance
Tc-99m sestamibi 140
made interpretation difficult. Tc-99m diethylenetriamine-
F-18 FDG 511
Cisternography
pentacetic acid (DTPA) and Tc-99m glucoheptonate,
In-111 DTPA 173, 247 renal radiopharmaceuticals, were subsequently used and
Tc-99m DTPA 140 did not have those problems. Uptake in the brain only
occurred if there was disruption of the BBB (e.g., with
tumor and stroke). The sensitivity for detecting disease
was quite good, ranging from 80% for stroke and 85% for
brain tumors. However, with the advent of CT in the
1980s, the use of BBB studies declined markedly.
Rolandic
Currently, functional nuclear medicine imaging stud-
Frontal fissure ies rely on PET and SPECT radiopharmaceuticals, which
lobe Parietal cross the BBB and reveal details not seen by CT or MRI.
lobe Of course, the anatomy seen on CT and MRI is much
Sylvian
fissure more detailed than the structures seen with SPECT or
PET, but many structures can be visualized ( Fig. 13-6).
Different categories of radiopharmaceuticals can be
Occipital used to evaluate regional cerebral perfusion ( rCBF ),
Temporal lobe brain metabolism (e.g., regional glucose metabolism,
lobe rCGM), or neurotransmitter receptor binding. Indica-
tions for nuclear medicine evaluation of the CNS are
listed in Box 13-2.
Cerebellum
Over a quarter of a century ago, PET imaging began
with the brain. Unlike other functional imaging tests
Insula such as functional MRI ( fMRI), PET offers the ability to
precisely measure or quantitate not only the one param-
eter being studied, but also global changes in flow and
metabolism. The physical properties of common PET
Figure 13-1 Cerebral cortex lobar anatomy. radionuclides used in the CNS are described in Table
13-1. F-18 fluorodeoxyglucose ( F-18 FDG) is the only
agent approved by the U.S. Food and Drug Admin-
istration ( FDA) and is available on a clinical basis.
However, an extensive array of radiopharmaceuticals
Radiopharmaceuticals targeting metabolism, blood flow, and neurotransmit-
Radiopharmaceuticals can be divided into those that do ters are under investigation. Some are listed in Box 13-3.
and those that do not cross the blood-brain barrier The clinical availability of PET technology is only
(BBBB).The first radionuclide brain scan agent, used in a recent phenomenon, and most clinical nuclear
 Central Nervous System 421

Anterior
cerebral

Middle
cerebral

Anterior Posterior
cerebral cerebral
Middle
Internal cerebral Basilar
Basilar carotid
Vertebral Internal
Posterior carotid
A carotid
B

Figure 13-3 Cerebral arterial anatomy. A, Coronal section shows circle

of Willis and course of the middle and anterior cerebral arteries. Internal
carotids divide at the base of the brain (circle of Willis) into the anterior and
middle cerebral arteries. The middle cerebral artery runs laterally in the
sylvian fissure, then backward and upward on the surface of the insula,
where it divides into branches to the lateral surface of the cerebral
hemisphere and to portions of the basal ganglia. B, Midline sagittal view
shows distribution of anterior, middle, and posterior cerebral arteries. The
anterior cerebral artery supplies the cerebrum along its medial margin
above the corpus callosum and extends posteriorly to the parietal fissure as
well as to the anterior portion of the basal ganglia. Vertebral arteries fuse
into the basilar artery, which branches at the circle of Willis into the two
Middle cerebral posterior cerebral arteries supplying the occipital lobe and the inferior half
and branches of the temporal lobe. C, Left lateral view shows distribution of the middle
C cerebral artery over the cerebral cortex.

Superior
sagittal
sinus
Lateral view Vertex view
Inferior
sagittal Anterior cerebral
sinus
Middle
Straight cerebral
sinus

Confluence
of sinuses
Transverse sinus Posterior
cerebral
Occipital sinus
Sigmoid sinus
Superior petrosal sinus Anterior view Posterior view
Cavernous
sinus Inferior Anterior cerebral
petrosal Internal jugular vein
sinus Middle
Figure 13-4 Cerebral venous anatomy. The superior sagittal cerebral
sinus runs along the falx within the superior margin of the Posterior
interhemispheric fissure. The inferior sagittal sinus is smaller, cerebral
courses over the corpus callosum, and joins with the great vein of
Galen to form the straight sinus, which drains into the superior
sagittal sinus at the confluence of sinuses (torcular Herophili) at
the occipital protuberance. Transverse sinuses drain the sagittal Figure 13-5 Regional cerebral cortex perfusion of the anterior,
and occipital sinuses into the internal jugular vein. middle, and posterior cerebral arteries.
422 NUCLEAR MEDICINE: THE REQUISITES

Figure 13-6 Normal distribution of F-18 FDG. High-resolution (A) transverse PET images with
corresponding levels on T1-weighted MRI (B).
Continued
 Central Nervous System 423

Figure 13-6, cont’d Coronal PET (C) and comparable T1-weighted MRI (D).F, Frontal lobe;T, temporal
lobe;P, parietal lobe;O, occipital lobe;C, cerebellum;Th, thalamus;Ca, caudate;PA, putamen.
424 NUCLEAR MEDICINE: THE REQUISITES

Box 13-2 Clinical Indications for Cortical Table 13-1 Positron Emission Tomography
Cerebral Scintigraphy Radionuclides

Dementia Imaging Maximum

Alzheimer’s disease Half-life photo beta-energy
Radionuclide (min) peak (keV ) (MeV )
Lewy body disease
Pick’s disease Fluorine-18 110 511 0.635
Multi-infarct dementia Carbon-11 20 511 0.970
Acquired immunodeficiency syndrome–dementia Nitrogen-13 10 511 1.2
complex Oxygen-15 2 511 1.7
Epilepsy
Stroke
Transient ischemia attacks
Head trauma
Movement disorders
Box 13-3 Positron Emission Tomography
Parkinson’s disease
Huntington’s chorea Radiopharmaceuticals
Psychiatric disorders
Attention deficit disorder Compound Application
Obsessive-compulsive disorder
Schizophrenia O-15 H2O Blood flow
Brain death O-15 O2 Oxygen metabolism/flow
Tumor imaging O-15 or C-11 Blood volume
carboxyhemoglobin
C-11 methionine Amino acid metabolism
C-11 methylpiperone Dopamine receptor activity
C-11 carfentanil Opiate receptor activity
C-11 flunitrazepam Benzodiazepine receptor
medicine brain studies utilize SPECT gamma camera tech- activity
nology. SPECT radiopharmaceuticals are lipophilic, C-11 scopolamine Muscarinic cholinergic
cross the intact blood–brain barrier, and are ideally receptors
labeled with Tc-99m. There is no SPECT radiotracer C-11 ephedrine Adrenergic terminals
F-18 fluorodeoxyglucose Glucose metabolism
equivalent to F-18 FDG for measuring metabolism.
( FDG)
SPECT agents generally reflect blood flow changes F-18 fluoro-L-dopa Presynaptic dopamine
because cortical rCBF is closely linked to the oxygen system
demands of the brain and usually approximates metab- F-18 fluorothymidine DNA synthesis
olism. Areas with more synaptic activity require (FLT )
greater blood flow. Activational studies can therefore
target areas of the brain showing increased flow when
activated by a certain task. In many ways, the SPECT
images are similar to the F-18 FDG PET images. Like
F-18 FDG, a similar pattern of increased gray matter to
white matter differential is seen. with PET, arterial sampling and even sometimes arterial
injections are needed for true quantitation.
Cerebral Blood Flow Radiopharmaceuticals The noble gas xenon-133 ( Xe-133) is best known as
Studying rCBF patterns can help diagnose disease. Many a ventilation lung scan agent. However, once in the
inert, freely diffusible radiotracers have been developed, lungs, it rapidly travels throughout the body and readily
which closely approximate rCBF as measured by micro- diffuses across the BBB. Given its high extraction effi-
sphere injection techniques. The positron emitter O-15 ciency, the perfusion pattern correlates better with
has long been used to measure rCBF in the form of O-15 actual cerebral blood flow measured by microsphere
labeled water ( O-15 H2O). Inhaled O-15 can be used to techniques than the other currently available SPECT per-
measure oxygen extraction and metabolism. Many tech- fusion agents. It is cleared quickly from the body, allow-
nical demands are created by the short half-life of O-15 ing for multiple acquisitions as might be needed in
(2 minutes), including rapid imaging, continuous radio- activational studies. However, this rapid clearance also
tracer input flow, and an onsite cyclotron. It should also makes Xe-133 technically demanding to use. Another
be noted that although accurate quantitation is possible limitation of Xe-133 is the lower- energy gamma photons

( 80 keV) and a high scatter fraction, which both
contribute to poor image quality. Xe-133 is a valuable
research tool, providing an inexpensive and noninvasive
approach to quantitate rCBF.
The lipophilic amines derivative iodine-123 isopropyl
iodoamphetamine ( I-123 IMP or Spectamine) localizes in
the brain by temporarily binding to amphetamine recep-
tors. With an extraction ratio greater than 90%,I-123 IMP
reasonably approximates rCBF. Imaging must be done
rapidly, usually within 1 hour, as it is not fixed in the neu-
rons but washes out. This allows for interesting imaging
possibilities, as abnormal but viable ischemic regions will
show “redistribution” like a thallium-201 cardiac stress
test ( Fig. 13-7). No longer clinically available, I-123 IMP
has been replaced by the Tc-99m agents, which have
better dosimetry and imaging characteristics.
Technetium-99m rCBF Radiopharmaceuticals
Currently, the neutral, lipophilic Tc-99m labeled hexa-
methylpropyleneamine oxime ( Tc-99m HMPAO) and
ethyl cysteinate dimer ( Tc-99m ECD) are used clini-
cally. Favorable characteristics of these two agents
include high first-pass extraction across the BBB, close
correlation with rCBF, and desirable 140 keV gamma
photons. However, both slightly underestimate true
rCBF, especially at high flow states. Unlike O-15 water,
Xe-133, and I-123 IMP, the Tc-99m perfusion agents are
relatively fixed in the neuron. Therefore, delayed imag-
ing shows what the perfusion pattern looked like at the
time of injection. If injected during an epileptic
seizure, perfusion abnormalities can be seen hours
after seizure resolution.
In practice, the distribution of Tc-99m HMPAO and
Tc-99m ECD is slightly different might be difficult to
compare directly in serial patient studies. Tc-99m
HMPAO accumulates more in the frontal lobes, thalamus,
and cerebellum, whereas Tc-99m ECD shows higher
affinity for the parietal and occipital lobes. Most clini-
cians use the agent with which they are most familiar.
Tc-99m HMPAO ( Tc-99m Exametazime or Ceretec)
was first introduced in the mid-1980s. Originally avail-
Figure 13-7 Type I-VI are described in Table 13-5.
Central Nervous System 425
able as a kit requiring use within 30 minutes of radiolabel-
ing, stabilizers have since been added allowing a 4-hour
shelf-life following the addition of the radiolabel. Tc-99m
HMPAO has a good first-pass extraction of roughly 80%,
with 3.5–7.0% of the injected dose localizing in the brain
within 1 minute of injection. Once across the BBB, it
enters the neuron and becomes a polar hydrophilic mole-
cule remaining trapped inside the cell. However, some of
the radiopharmaceutical may be present in different iso-
meric forms, which do not remain trapped. Although up
to 15% of the dose washes out in the first 2 minutes,there
is little loss over the next 24 hours. SPECT image acquisi-
tion can be done anywhere from 15 minutes to 2 hours
after injection. Excretion is largely renal ( 40%) and gas-
trointestinal ( 15%).
Tc-99m ECD ( Tc-99m Bicisate, Neurolite) is a neutral
lipophilic agent that passively diffuses across the BBB
like Tc-99m HMPAO. Once prepared, the Tc-99m ECD
dose is stable for 6 hours. Like Tc-99m HMPAO, it also
underestimates true rCBF. It has a first-pass extraction
of 60–70%, with peak brain activity reaching 5–6% of
the injected dose. The blood clearance is more rapid
than Tc-99m HMPAO, resulting in better brain-to-back-
ground ratios. At 1 hour, less than 5% of the dose
remains in the blood, compared to over 12% of a Tc-99m
HMPAO dose.
Once inside the cell, the mechanism of Tc-99m ECD
retention differs from Tc-99m HMPAO, as it involves enzy-
matic de-esterification forming polar metabolites unable
to cross the cell membrane. However, there is a slow
(roughly 6% per hour) washout of some labeled metabo-
lites. Although images may be superior to Tc-99m
HMPAO 15–30 minutes after injection, they may be sub-
optimal if imaging is delayed. Almost 25% of the brain
activity has cleared if imaging is delayed 4 hours.
Glucose Metabolism––F-18 FDG
The brain is an obligate glucose user, and F-18 FDG is
a glucose analog allowing accurate assessment of regional
glucose metabolism (rCGM). F-18 FDG is able to cross
the blood–brain barrier utilizing glucose transporter sys-
tems and enters the neuron. After rapid phosphorylation
by hexokinase-1, F-18 FDG is metabolically trapped and
cannot proceed further along the glucose metabolism
pathway. F-18 FDG PET, with a resolution of 4–5 mm, is
superior to the 7 mm of SPECT. Approximately 4% of the
administered dose is localized to brain. By 35 minutes
after injection, 95% of peak uptake is achieved. Urinary
excretion is rapid with 10–40% of the dose cleared in
2 hours. In addition to reflecting rCBF, as a marker of glu-
cose metabolism, F-18 FDG can be used to determine
tumor viability. With its 110-minute half-life, F-18 FDG
does not require an expensive onsite cyclotron. As PET
technology enters mainstream medicine, costs (including
the price of the dose) continue to drop.
426 NUCLEAR MEDICINE: THE REQUISITES
Normal Distribution of rCBF and rCGM Agents
As blood flow relates to brain activity, the greater than
2:1 differential in blood flow of gray matter to white mat-
ter seen on SPECT and PET scans is not surprising.
Lesions in the white matter often can not be detected or
even differentiated from cerebrospinal fluid (CSF)
spaces on PET or SPECT. Close correlation with MRI or
CT is necessary for identifying white matter changes and
enlarged ventricles.
It is important to be familiar with normal radiophar-
maceutical distribution and the factors that can alter
that pattern. Typically, activity is fairly evenly distrib-
uted between the lobes of the brain. However, this is
dependent on the conditions at the time of injection.
For example, bright lights will increase occipital lobes
activity, falsely causing the frontal lobes to appear
decreased. Therefore, conditions must be carefully con-
trolled. Having intravenous access in place well before
the dose is injected helps decrease stimuli. The patient
should be in a quiet, dimly lit room, with eyes usually
kept open.
The normal distribution of SPECT and PET agents also
changes with age. In children, there is a relative decrease
in perfusion to the frontal lobes. This increases over time,
reaching an adult level by about 2 years of age. In adults,
global activity decreases with age, and this decrease is
more prominent in the frontal regions. Utilizing compari-
son age-matched normal databases and computer
programs that quantitate rCBF may help improve not only
sensitivity but specificity given these changes.
Methodology
Imaging protocols will vary greatly depending on the
specific instrumentation used. However, it is safe to say
that when SPECT is being performed in the brain, more
heads are better than one. Dedicated triple-head gamma
cameras are increasingly rare but yield the best results if
available. A dual-headed camera creates images superior
to a single-headed SPECT camera. Patient positioning is
just as important as the equipment used. The heads of
the camera must come as close to the patient as possible
or resolution is lost. In heavy patients and those whose
shoulders get in the way, the posterior fossa may not be
seen. A protocol for Tc-99m cerebral perfusion SPECT
imaging is given in Box 13-4. The SPECT images may be
processed using filtered back projection, although newer
systems offer iterative reconstruction. A filter is applied
to smooth the image. In general, filters can be sampled
and modified for each patient in the postprocessing stage
to achieve an optimal image.
PET imaging with F-18 FDG provides higher resolu-
tion than SPECT imaging with the technetium-labeled
perfusion radiopharmaceuticals. Resolution can be
optimized by acquiring images with septa in front of
Box 13-4 SPECT Cerebral Perfusion
Imaging: Protocol Summary
PATIENT PREPARATION
None.
RADIOPHARMACEUTICAL
20 mCi technetium-99m HMPAO (Ceretec) or Tc-99m
ECD (Neurolite)
INSTRUMENTATION
Camera: triple-headed SPECT or dual headed SPECT
Collimators: ultra-high resolution parallel hole
Computer setup: SPECT acquisition parameters
Matrix size: 64 × 64
Zoom: 2
Rotation: step and shoot
Orbit: circular
Angle step size: 3 degrees
Stops: 40 per head
Time per stop: 40 sec (total time, 1600 sec or 27 min)
IMAGING PROCEDURE
Prepare dose according to package insert. Note shelf life.
Begin IV or tape butterfly in place; make patient
comfortable in quiet, dimly lit room; inject; eyes open;
begin scanning in 15 minutes to 2-3 hours later
Position patient so that brain is entirely within field of
view of all detectors
Position collimators as close as possible to patient’s head
Record radius
PROCESSING
Filtered back projection or iterative reconstruction
Filter: Hamming, 1.2 high-frequency cutoff
Attenuation correction: 0.11 cm−1
the detectors (two-dimensional [2-D] mode). The
septa minimize scatter. Many sites prefer the higher
count rate three-dimensional mode, which does not use
septa. Although resolution is lowered by scatter, many
camera systems have done away with 2-D mode for the
brain. A sample protocol is listed in ( Box 13-5). Like
SPECT studies, the patient should be injected in a quiet,
dimly lit room. The patient should remain undisturbed
during the uptake period. Postprocessing protocols
typical use iterative algorithms rather than filtered back
projections.
Dosimetry
The dosimetry of cerebral radionuclide imaging depends
on the radiopharmaceutical utilized. A comparison of
several agents is given in Table 13-2.
 Central Nervous System 427

Box 13-5 PET Imaging Protocol CLINICAL APPLICATIONS OF

CEREBRAL PERFUSION IMAGING
PATIENT PREPARATION
Dementia
Fast 4-6 hours, avoid carbohydrates, maintain normal
blood glucose. As our population ages, dementia has an increasing
Check blood glucose. If <180–200, continue. impact on our society and healthcare systems. Patients
Elevated glucose: consider rescheduling, administer may display differing clinical symptoms (e.g., short- or
insulin, recheck, delay 2 hours. long-term memory loss, loss of judgment, personality
Inject in quiet, dimly lit room, eyes open. changes, and loss of other higher cortical functions).
Wait 35-45 minutes The functional decline can occur rapidly over months or
RADIOPHARMACEUTICAL slowly over years. The mental changes must be differen-
tiated from the normal decline in memory and decreased
5-15 mCi (185–555 MBq) F-18 fluorodeoxyglucose (F-18
FDG) ability to learn new things that accompany aging.
However, the clinical diagnosis is often difficult and
INSTRUMENTATION delayed, and MRI may not reveal changes such as atrophy
Dedicated PET or PET-CT camera until the end stages of disease.
Coincidence camera Dementia is a manifestation of many diseases, not all
primarily of the brain ( Box 13-6). Only about 10% of
IMAGING
dementias, such as those caused by vitamin B12 or thy-
Attenuation correction scan: gadolinium rods or CT roid hormone deficiency, are treatable. However, with
Acquisition: 7 min per bed position for one bed position better understanding of the diseases underlying demen-
PROCESSING tia, new therapies are being developed. For example, the
cholinesterase inhibitors have shown some stabilizing
Iterative reconstruction, automated software
effects in patients with AD, occasionally even reversing
perfusion trends seen on SPECT scans. As treatments are
developed, early diagnosis will likely be more critical.
SPECT and PET can help diagnose the etiology of
dementia much earlier than clinical criteria or MRI.
Studies have shown that combining clinical diagnosis

Table 13-2 Absorbed Radiation Doses

Tc-99m HMPAO Tc-99m ECD F-18 FDG Tl-201

Dose Rads/20 mCi Rads/20 mCi Rads/10 mCi Rad/2mCi

(cGy/740 MBq) (cGy/740 MBq) (cGy/370 MBq)

Brain 0.5
Lens and retina 0.5
Heart 1.5 1.0
Lung 0.2 0.6
Liver 1.1 0.2 0.6 1.1
Spleen 1.4
Gallbladder 3.8 4.0
Kidney 2.6 0.5 0.7 2.4
Large bowel 1.6 1.1
Bladder 0.9 2.2 4.1
Testes 0.1 0.3 1.0
Ovaries 0.5 0.6 1.0
Red marrow 0.2
Total body 0.3 0.2 0.4 0.4
428 NUCLEAR MEDICINE: THE REQUISITES
Box 13-6 Causes of Dementia
Alzheimer’s disease
Parkinson’s disease
Lewy body disease
Multi-infarct dementia
Pick’s disease
Progressive supranuclear palsy
Creutzfeldt-Jacob disease
Huntington’s chorea
Multiple sclerosis
Vitamin B12 deficiency
Endocrine disorders (hypothyroidism)
Chronic infection (tuberculosis, syphilis)
Human immunodeficiency virus (HIV) encephalopathy
Alcohol and drugs
with perfusion findings on SPECT raises the positive pre-
dictive value compared to either method alone. PET has
equal or better results in comparison with SPECT. Some
investigations report that PET has a sensitivity and speci-
ficity of 94% and 73%, respectively.
Although PET has higher sensitivity and higher resolu-
tion than SPECT, the overall patterns seen in disease are
similar for both rCGM and rCBF. Certain perfusion pat-
terns have been recognized that are characteristic for dif-
ferent types of dementia. In general, these types of
dementia can be characterized as posterior, frontotempo-
ral, or vascular. The posterior dementias include AD, Lewy
body disease, and dementia of PD. The frontotemporal
dementias include Pick’s disease and primary progressive
aphasia. There is some overlap of vascular dementia and
frontotemporal dementia. Multi-infarct dementia tends to
display a generalized decrease in activity.
Posterior Dementias
AD is the most common of the dementias. It is estimated
that nearly 10% of the population over 65 and 50% of
those over 85 are affected. Diagnosis can be definitively
made at autopsy or by the rarely performed brain biopsy
with pathological samples showing characteristic neu-
rofibrillary tangles and amyloid plaques. Originally
described as a dementia of a relatively young person
( presenile dementia), it is now recognized that many
older people originally thought to have multi-infarct
dementia actually have AD. On the other hand, 25% of
those thought to have AD clinically were found to have
other etiologies at autopsy.
Imaging findings vary with the stage of disease,but char-
acteristic patterns are well established in AD for both PET
and SPECT. Early disease tends to involve the superior pari-
etotemporal cortex manifested as hypometabolism/hypo-
perfusion (Fig. 13-8A). Although the classic examples are
symmetric, AD is often rather asymmetric,especially in the
early stages ( Fig.13-8B). As the disease progresses, it
involves the frontal cortices, although parietal and tempo-
ral lobe involvement usually remains greater. It may be
more difficult to diagnose very elderly patients and end-
stage patients because the imaging pattern may be more of
a generalized nonspecific decrease in cortical uptake
( Fig. 13-8C). However, sparing of the occipital visual cor-
tex, somatosensory and motor cortex, basal ganglia, thala-
mus,and cerebellum is the norm (Fig.13-9).
The strength of FDG PET may be in differentiating
which patients with so-called mild cognitive impairment
( MCI) will progress on to actual AD. MCI patients have
deficits on mini-mental status testing but do not
meet probable AD criteria. Approximately 15% of MCI
patients progress to AD per year. PET studies have
shown that MCI patients with parietal and temporal
hypoperfusion were much more likely to go on to AD
than those without the pattern. Identification of patients
in the presymptomatic or early phases of AD will be cru-
cial for therapy in the future.
New PET radiopharmaceuticals are being investigated
that bind to amyloid, muscarinic receptors, nicotinic
receptors, and components of the cholinergic system.
Correlation with genetic factors influencing AD will be
ongoing. Recent developments have shown a strong
link between scintigraphic perfusion patterns and cer-
tain genes, including the E4 allele of the apolipoprotein E
gene in AD. The other diseases that affect the posterior
regions of the brain often overlap with each other and
with AD. These include Lewy body disease ( dementia
with Lewy bodies or DLB) and Parkinson’s disease ( PD).
Roughly 30% of PD patients will develop dementia.
Diagnosis of the dementia may be complicated clinically
by depression. DLB is second only to AD among
dementias caused by neurodegenerative disorders
rather than by vascular etiologies. Pathological speci-
mens of PD patients show Lewy body intracellular
inclusions in the midbrain. In DLB, these Lewy bodies
are found in the cerebral cortex. Clinically, these DLB
patients often demonstrate a fluctuating dementia, falls,
some Parkinsonian symptoms, such as tremor and visual
hallucinations. It is likely that DLB and PD are related
as part of a spectrum of disease. Movement disorder
patients who later develop dementia are considered to
have PD, whereas DLB patients are those who develop
the movement disorder at the same time or later than
dementia. FDG PET and SPECT perfusion agents
images can confirm DLB by showing changes in the
posterior cortical regions. The pattern is often simi-
lar to AD but tends to involve the occipital lobes to
a greater extent ( Fig. 13-10). The involvement of the
primary visual cortex can explain the visual hallucina-
tions. Another deviation from the AD patterns may be
preservation of hippocampal activity in DLB but with
 Central Nervous System 429

Figure 13-8 Alzheimer’s disease. A, Transaxial PET images reveal decreased perfusion to the
temporal parietal cortex beginning high in the posterior parietal region with sparring of the basal
ganglia, thalamus and cerebellum. B, Tc-99m HMPAO SPECT show that while Alzheimer’s is
classically described as a symmetrical process, it may be quite asymmetric, as seen in the left
posterior parietal region.
430 NUCLEAR MEDICINE: THE REQUISITES

Figure 13-8, cont’d C, As Alzheimer’s becomes more severe it is more diffuse on this
SPECT study.

Figure 13-9 Alzheimer’s disease on PET-CT. A patient with

moderately advanced Alzheimer’s shows hypometabolism not only
in the posterior parietal and temporal regions but involvement
now extends anteriorly. Sparring of the occipital region and
sensory motor cortex is clear (arrowheads).
possibly greater involvement of the cerebellum. The
imaging pattern of dementia in PD is also similar to AD
with the exception of greater occipital involvement and
more mesial temporal sparing. As depression (which
often occurs in PD) can be confused with dementia, it is
important to note these depressed PD patients can show
decreased prefrontal and caudate activity rather than the
posterior pattern so typical of dementia. However, other
patterns have been described.
Figure 13-10 Comparison of posterior dementias on Tc-99m
HMPAO. Top row: Alzheimer’s disease generally begins near the
superior convexity and involves the parietal and temporal regions
laterally (arrows), sparing the occiput and cerebellum
(arrowhead ). Bottom row: Lewy body disease involves the
medial occipital region (arrows) and usually has more caudal
extension than Alzheimer’s.
Frontotemporal Dementia
The frontotemporal dementias ( FTD) are a diverse group
of diseases that involve the frontal and temporal lobes
( Fig. 13-11). Clinically, patients show varying presenta-
tions. Aphasia occurs with temporal lobe abnormalities
 Central Nervous System 431

Figure 13-11 Frontotemporal dementia sparing posterior parietal regions. A, Frontal

hypometabolism on PET can be due to many causes and changes visible long before MR shows
atrophy or signal changes. B, Post-contrast T1-weighted MR shows no atrophy and other MR
sequences were unremarkable.
432 NUCLEAR MEDICINE: THE REQUISITES
Figure 13-11, cont’d C, Pick’s disease is rare. It classically
shows frontal hypoperfusion with a sharp anterior to posterior
cutoff, but the findings are somewhat nonspecific and other causes
for decreased frontal activity must be considered.
and frontal lobe involvement results in personality
changes, including loss of judgment and inappropriate
behavior. In FTD, memory loss is often secondary or
absent as opposed to being the primary problem as in
AD. The differential of FTD includes Pick’s disease,
semantic dementia, primary progressive aphasia, and
familial FTD. Pick’s disease is the classic but rare FTD
showing frontal and anterior temporal neuronal degener-
ation ( Fig.13-11C). Pick bodies are sometimes found but
amyloid plaques, neurofibrillary tangles and Lewy bodies
are absent. Severe atrophy on MRI is not visualized until
much later than the perfusion and metabolic changes of
SPECT and FDG PET.
Decreased perfusion and metabolism in the frontal
and temporal lobes may be seen in diseases other than
those in the FTD group. These include cocaine abuse,
depression, progressive supranuclear palsy, spinocerebel-
lar atrophy, amyotrophic lateral sclerosis, and very rarely
in AD as an isolated finding. More commonly, decreased
frontal and temporal rCBF and rCGM is secondary to
a vascular process. Focal cortical defects, abnormalities
on MRI, and areas of scintigraphic asymmetry should
raise the level of suspicion for a vascular degenerative
process.
Vascular Dementia
In vascular dementia, MRI often shows changes such as
focal white matter lesions (subcortical encephalomala-
cia) and tiny cortical infarcts, which may not be visible
on a SPECT or FDG PET study. However, because AD can
overlap with vascular etiologies, PET or SPECT may be
ordered in a patient with nonspecific vascular changes
on MRI or with multi-infarct dementia. A strong frontal
predisposition may be present in vascular dementia
which must be differentiated from expected age-related
decreases. Often, the generalized decrease in rCGM or
rCBF seen in the vascular dementia patient is difficult to
differentiate from severe Alzheimer’s ( Fig. 13-12).
Sometimes small disruptions in the cortical ribbon or
subcortical defects such as in the thalamus can be seen.
At other times, a regional decrease is seen that clearly
correspond to regions supplied by main arterial
branches such as the MCA. At times, when other causes
for dementia such as Alzheimer’s disease are excluded,
vascular dementia may be left as a diagnosis of exclusion.
HIV Encephalopathy and AIDS-Dementia Complex
Early clinical signs of human immunodeficiency virus
( HIV) and acquired immunodeficiency syndrome ( AIDS)
are frequently subtle and may be difficult to distinguish
from depression, psychosis, or focal neurological disease.
Because treatment (e.g., with AZT ) can improve cogni-
tive function, early detection is helpful. Findings on CT
and MRI are not specific.
Cerebral perfusion SPECT is highly sensitive for AIDS-
dementia complex and shows a typical scintigraphic pat-
tern of patchy,multifocal cortical and subcortical regions of
hypoperfusion. These changes occur most frequently in
the frontal,temporal,and parietal lobes,although basal gan-
glia involvement is common. Patients may also have focal
areas of increased activity. The perfusion pattern can
improve with therapy. A similar brain perfusion pattern
has been described in chronic cocaine and polydrug users.
Stroke and Cerebrovascular Disease
Although MRI and CT are now preeminent modalities for
diagnosing stroke, there is a renewed interest in assessing
patients with cerebrovascular disease with PET and SPECT
to determine the role of therapeutic intervention. Nuclear
medicine studies are frequently positive earlier than CT
and may show characteristic cortical defects (Fig. 13-13).
Functional imaging offers the ability to visualize and quanti-
tate rCBF and rCGM, which can assess parameters of dis-
ease not seen on structural imaging. PET and SPECT can
help determine which patients are at risk for stroke, most
likely to benefit from intervention, and even predict stroke
recovery. Relationships such as distant neuronal activity
loss (diaschisis),neuron recruitment,and recovery through
neuronal plasticity can be studied (Fig.13-14).
The role of functional imaging in the acute stroke is
evolving. The temporal development of a stroke has
been studied. In the very early hours, tissue oxygen
metabolism is maintained in the face of markedly
reduced rCBF. This mismatch or “misery perfusion”
implies viable tissue could be rescued, and the area at
risk may be larger than seen on MRI. This is especially
important in the periphery, or penumbra, of the
affected area. PET and SPECT may play a part in deter-
mining who should get thrombolytic therapy, predict-
ing recovery, and determining the amount of tissue at
risk. Eventually, as the ischemia progresses, damage to
vessels may result in excessive or “luxury perfusion,”
 Central Nervous System 433

Figure 13-12 Vascular dementia, often shows a frontal predisposition similar to other FTD.
A, Vascular dementia may be more diffuse as seen in this patient and difficult to differentiate from
other etiologies including severe Alzheimer’s. B, Slow progression was seen in this patient clinically
and on a repeat scan 5 years later.

Figure 13-13 Left parietal middle cerebral artery cortical stroke. HMPAO SPECT shows
significantly decreased perfusion to the region including subcortical structures.
434 NUCLEAR MEDICINE: THE REQUISITES
Figure 13-14 Crossed cerebellar diaschisis. Top row, A sub
cortical stroke in the left parietal region on CT is subtle on
SPECT as the gray matter where HMPAO localizes is not heavily
involved. No abnormality is seen in the cerebellum on CT.
Bottom row, One type of distant stroke effect not seen on CT is
revealed on the Tc-99m HMPAO SPECT. Hypoperfusion of the
contralateral cerebellum or crossed cerebellar diaschisis is
seen.
obscuring the early findings seen with Tc-99m
HMPAO.
Carotid artery balloon occlusion studies are well
established in the assessment of vascular reserve.
These studies are used in patients who may require
permanent internal carotid artery occlusion such as for
treatment of intracranial aneurysm. An intravenous
injection of Tc-99m HMPAO is done in the angiography
suite at the time of internal carotid artery balloon
occlusion. The balloon is deflated after 1 minute.
Imaging is done once the catheter has been removed.
A cerebral arteriogram can show which patients have
an intact Circle of Willis and cross-filling. WADA neuro-
logical testing is also sometimes possible during the
temporary occlusion. Amobarbitol infused into the
carotid is used to predict speech and memory func-
tion. However, up to a 20% decrease in morbidity and
mortality can be seen by adding a Tc-99m HMPAO
SPECT scan to the preoperative workup. Patients at
risk for stroke following a planned permanent arterial
occlusion are easily identified with a significant drop in
perfusion to the occluded side on SPECT images.
Carotid bypass is generally warranted in these patients
as the remaining vessels can not meet the needs of the
side that will be occluded ( Fig. 13-15).
Figure 13-15 Left, HMPAO SPECT performed after a carotid
balloon occlusion injection shows severe left middle cerebral
artery territory hypoperfusion identifying a high risk for stroke.
The patient underwent permanent left carotid occlusion using
gradual occlusion with a Selverstone clamp rather than bypass.
Right, This gradual occlusion was not sufficient to protect the
patient and the patient did indeed suffer a stroke seen on CT in the
same distribution as the SPECT.
SPECT can help assess suspected ischemia and
stroke risks in the cases of transient ischemic attacks
( TIAs) and vascular diseases like atherosclerosis and
Moyamoya. Previously, I-123 Spectamine was used to
show reversibility of a region of ischemia ( Fig. 13-7).
The vascular reserve of patients can be assessed by
a pharmacologic stress test by imaging after vasodilatory
response to increased CO2 caused by acetazolamide
( Diamox), a carbonic anhydrase inhibitor and antihyper-
tensive agent.Vasodilation following intravenous admin-
istration of 1 gram of Diamox leads to an increase in
rCBF. Although global blood flow is increased, abnormal
vessels cannot dilate and blood is shunted away. This
will accentuate any abnormality and better demonstrate
territories at risk for infarction ( Fig. 13-16). It is impor-
tant to compare the resting state to the stress state.
Change may be best assessed with semiquantitative
analysis of cortical activity.Xe-133 may have superior sen-
sitivity to Tc-99m HMPAO, although it is not frequently
used clinically.
Brain Death
Diagnosis of brain death is primarily clinical. Accuracy
and speed in making the diagnosis become critical when
organ donation is considered and life support systems
must be used. Clinical diagnosis may be difficult due to
the specific criteria necessary to make the diagnosis of
brain death, as follows:
1. The patient must be in deep coma with total absence
of brainstem reflexes or spontaneous respiration.
2. Potentially reversible causes such as drug intoxica-
tion, metabolic derangement, or hypothermia must
be excluded.
3. The cause of the brain dysfunction must be diag-
nosed (e.g., trauma, stroke).
 Central Nervous System 435

Figure 13-16 Semiquantitative analysis of ischemia with Diamox SPECT. A, Tc-99m HMPAO
SPECT before Diamox (left) shows mild left parietal hypoperfusion. After Diamox (right), the brain
shows increased activity in response to vasodilation with the exception of the left parietal defect
and the decreased activity from ischemia is even more apparent. B, Cortical ROI boxes are drawn
like a clock numbered from 1 o’clock to 12 o’clock to generate rCBF curves. In this patient, the
post-Diamox curves are increased relative to rest perfusion with the exception of a dip from 2
o’clock to 4 o’clock. This signifies an area at high risk for infarction because it is unable to respond
to the vasodilation.

4. The clinical findings of brain death must be present
for a defined period of observation ( 6–24 hours).
Confirmatory ancillary tests are used by clinicians to
increase certainty, but the diagnosis of brain death is still
primarily a clinical one. An isoelectric electroencephalo-
gram (EEG) by itself does not establish brain death, and at
least one repeat study is required. In the patient with
intoxication from barbiturates and other depressive drugs
or with hypothermia, the EEG may be flat, even though
cerebral perfusion is still present and recovery is possible.
Lack of blood flow to the brain is diagnostic of brain
death. Edema, softening, necrosis, and autolysis of brain
tissue lead to increased intracranial pressure. As pres-
sure rises, eventually it prevents intracranial perfusion.
This can be demonstrated with four-vessel arteriography,
but the test is invasive and unnecessary.
The radionuclide brain death study is usually per-
formed when the EEG and clinical criteria are equivocal.
It is simple, rapid, and can be performed at the bedside.
Scintigraphy is not affected by drug intoxication or
hypothermia. An abnormal radionuclide angiogram
showing no cerebral perfusion is more specific for brain
death than an isoelectric EEG.
Radiopharmaceuticals
Brain death can be diagnosed using a radionuclide flow
study alone because the lack of intracerebral blood flow
is diagnostic. Technetium-labeled radiopharmaceuticals
are used to assess dynamic flow. Tc-99m DTPA was
often used because it is cleared rapidly from the blood,
allowing a repeat study if necessary. However, optimal
technique is mandatory and interpretation may be diffi-
cult. Visualization of normally draining veins can be
helpful ( Fig. 13-17).
Tc-99m HMPAO and Tc-99m ECD studies are easier
to interpret and are now preferred ( Fig. 13-18). Flow
images can be obtained but are not necessary, because
delayed images will show the fixed presence or absence of
brain uptake,requiring flow to the brain. If no CBF is pres-
ent, no cerebral uptake will occur. Planar images are ade-
quate,and SPECT is not necessary to diagnose brain death.
Methodology
A scalp tourniquet has been suggested for studies utilizing
Tc-99m DTPA to minimize flow through the external
carotid arteries, facilitating image interpretation of brain
perfusion. However, a tourniquet is contraindicated in
children, because it could increase intracranial pressure.
Adults also probably do not require a tourniquet, because
peripheral scalp activity can be differentiated from cere-
bral perfusion. An adequate radiopharmaceutical dose of
10–20 mCi (370–740 MBq) and good bolus are required
to ensure a diagnostic flow study. Images of the injection
site can be performed to ensure the dose was adequate
and not infiltrated. The radionuclide angiogram protocol
for CBF is used in brain death (Box 13-7).
Image Interpretation
Diagnostic findings of brain death include the lack of
intracranial arterial flow or major venous sinuses on
subsequent static images ( Tc-99m DTPA). Flow to
both common carotid arteries is seen to the level of
the base of the skull. No visualization of the brain is
seen with Tc-99m HMPAO. Often the “hot nose” sign is
seen as increasing intracranial perfusion diverts
436 NUCLEAR MEDICINE: THE REQUISITES

Figure 13-17 Normal delayed Tc-99m DTPA planar images. (A) Anterior, (RL) right lateral, (LL)
left lateral, and posterior projections. The superior sagittal sinus (1) is seen on anterior and
posterior views. The floor of the frontal sinus (2), confluence of sinuses (3), transverse sinuses (4),
and sphenoid sinus (5) are faintly seen.

intracranial blood flow into external carotid circula-
tion, resulting in relatively increased flow to the face
and nose. This pattern is nonspecific and can also be
seen in internal carotid artery occlusion without brain
death. Faint visualization of the venous sagittal or
transverse sinus in the absence of intracranial perfu-
sion is also sometimes seen on Tc-99m DTPA scans.
Due to such factors as variable anatomy, delayed
images may be difficult to interpret. Furthermore,
flow images may be inadequate due to poor bolus
technique or computer malfunction. There-fore, Tc-
99m DTPA is clearly less desirable than Tc-99m
HMPAO.
Epilepsy
Intractable or medically refractory seizures may require
surgery for therapy. Precise seizure localization often
requires a combination of scalp EEG, MRI, magneto-
encephalography ( MEG), and nuclear medicine imaging
for evaluation. These noninvasive studies are important
in directing the invasive intracranial EEG grid placement
in the operating room and determining therapeutic
options. Although MRI often reveals abnormalities at the
site of seizure foci, such as mesial temporal hippocampal
sclerosis, it is rare for structural imaging to fully visualize
the actual extent of the abnormally activated neurons by
structural imaging. EEG remains critical in seizure local-
ization, but is often inconclusive.
PET and SPECT have very important roles in such
seizure evaluation. In the ictal state, activated foci
show increased activity, representing increased rCBF
and glucose metabolism. Interictal images, however,
show normal or decreased activity. In the postictal
state, activity is changing and may show areas of
increased and decreased activity. Clinical knowledge of
the seizure status is essential and is best accomplished
with continuous monitoring. In addition, as patients
may have more than one type of seizure, it must be
determined if the seizure of interest was occurring dur-
ing the ictal state.
 Central Nervous System 437

Figure 13-18 Brain death evaluation. A, CT images of a head trauma victim show a right frontal
parenchymal hemorrhage extending into the ventricles. B, Tc-99m HMPAO exam to evaluate for
brain death. The exam does not show the early arterial phase of perfusion but good cortical
perfusion and venous drainage into the superior sagittal sinus and (C) good delayed cortical uptake
although with a defect from the hemorrhage. D, A follow-up study 3 days later shows internal
carotid arterial flow terminating below the head from brain death. E, This is confirmed with absent
delayed cortical uptake. Any peripheral activity is due to external carotid flow.
438 NUCLEAR MEDICINE: THE REQUISITES
Box 13-7 Brain Death Scintigraphy:
Protocol Summary
PATIENT PREPARATION
None.
RADIOPHARMACEUTICAL
Technetium-99m HMPAO (or Tc-99m DTPA), 20 mCi
(740 MBq)
INSTRUMENTATION
Collimator: high resolution, low energy. Camera setup:
large-field-of-view gamma.
Window: 15% over 140-keV photopeak
Camera formatter setup: 2- to 3-sec flow images for
30 sec, then immediate and delayed static images in
multiple views
Computer setup: 1-sec flow images for 60 sec
(64 × 64 byte mode), then static images
(128 × 128 frame mode)
IMAGING PROCEDURE
1. Inject radiopharmaceutical as an intravenous bolus.
2. Acquire dynamic flow study.
3. Immediate 750k static images in the anterior,
posterior, right lateral, and left lateral views
(optional). Image injection site.
4. Starting 2 hours after injection,acquire delayed 750k
static images in the anterior,posterior,right lateral,
and left lateral views. Vertex view if needed.
Although ictal studies are most sensitive, they are
technically highly demanding and done with SPECT
(Fig. 13-19). Patients must be admitted and continu-
ously monitored off medication. Once the seizure is
identified, trained personnel must inject the radio-
tracer within seconds of seizure onset. Although imag-
ing can then be delayed, the patient must be able to
cooperate with imaging in a reasonable amount of time.
Ictal PET would not be practical given the half-life of
F-18 FDG. Interictal studies are far less sensitive,
although interictal PET appears to be better than inter-
ictal SPECT. Clinical knowledge of the seizure is
needed to be sure that a study is truly ictal or interictal.
Ictal SPECT has a sensitivity of nearly 90% in temporal
lobe seizures, and the abnormal areas are generally
more extensive than any structural abnormality on MRI.
However, sensitivity for extratemporal seizures is much
lower, on the order of 50–75%. Interictal FDG PET
and SPECT is approximately 70% sensitive for seizure
localization.
Tumor Imaging
The SPECT agents Tc-99m HMPAO and Tc-99m ECD are
generally not useful for the detection of intracranial
malignancies. Of the two agents,Tc-99m ECD may occa-
sionally show increased activity. However, both agents
usually show normal or decreased activity. Although
MRI is the most sensitive method for detecting metastatic
lesions, there are several situations where MRI is limited
and nuclear medicine studies offer potential solutions.
The most common of these is the detection of recurrent
gliomas when MRI shows signal changes and enhance-
ment, which could be due to either radiation necrosis
and surgical therapy. The differentiation of intracranial
lymphoma from Toxoplasmosis in immunocompromised
patients is another situation where MRI is often inconclu-
sive as both processes can show ring-enhancing lesions.
F-18 FDG PET has long played an important role in
diagnosing recurrent gliomas when MRI shows nonspe-
cific posttherapy signal changes. Evaluating changes in
metabolism is a more specific way to assess intracranial
tumor compared to agents relying on blood flow and
BBB breakdown. Because more aggressive tumors have
higher metabolic activity,glucose metabolism is increased
( Fig. 13-20). F-18 FDG uptake is, therefore, related to
tumor grade. Therefore, PET can help direct biopsy and
grade tumors.
Primary tumors can be graded based on the amount of
F-18 FDG uptake. Low-grade gliomas ( WHO grade I and
II) typically show uptake similar to white matter, whereas
high-grade tumors ( WHO grade III ) are similar or
increased compared to grey matter. Grade IV ( glioblas-
toma multiforme) shows markedly increased activity
compared to normal cortical grey matter. Interestingly,
low-grade pilocytic astrocytomas and benign pituitary
tumors can show increased F-18 FDG accumulation.
Following therapy, gliomas commonly recur. Detection
of recurrent gliomas is a common problem clinically as
the MRI typically shows persistent increased T2 signal
changes and contrast enhancement. PET images show
absent or decreased activity in the normal postoperative
brain and any area of increased uptake most likely repre-
sents tumor. Direct side-by-side comparison with the
MRI is critical for image interpretation, and actual fusion
of PET images to the MRI is extremely helpful. In the
case of high-dose radiation therapy, increased F-18 FDG
activity can be seen and may persist. Although this activ-
ity is generally mild and not greater than normal cortical
uptake, serial images to look for any areas of increasing
activity may be necessary to exclude early recurrence.
Recurrences are often more aggressive with intense
radiotracer accumulation.
 Central Nervous System 439

Figure 13-19 Seizure imaging. A, Ictal HMPAO SPECT axial (top) and coronal (bottom) images
reveal increased perfusion to the right temporal region from an active seizure. B, The abnormal
temporal region is a subtle area of hypometabolism on the Interictal FDG PET. C, A second patient
ictal SPECT demonstrates hyperperfusion in the right parasagittal region (left) which is an area of
hypometabolism on interictal PET (right) from a seizure focus.
440 NUCLEAR MEDICINE: THE REQUISITES
Figure 13-20 Recurrent gliomas may be difficult to detect on
MRI.T2-weighted MRI shows posttherapy signal changes (left).
The recurrent tumor is seen as an intense focus on the FDG PET
(right).
Although FDG PET is a valuable clinical tool in the
workup of many types of malignancy outside of the CNS,
nearly two thirds of intracranial metastatic lesions are not
seen on PET due to the high background activity.
Therefore, MRI remains the standard for metastatic lesion
detection.
Numerous other PET radiopharmaceuticals have been
made which image aspects of metabolism other than glu-
cose, such as DNA synthesis, protein synthesis, and fat
metabolism. Although limitations often persist in low-
grade tumor evaluation, several agents such as F-18
thymidine and C-11 methionine demonstrate improved
sensitivity and specificity over FDG.
SPECT Tumor Imaging
SPECT evaluation of recurrent gliomas and differentiat-
ing intracranial lymphoma can be done with the cardiac
imaging agents thallium-201 ( Tl-201) and Tc-99m ses-
tamibi ( Fig. 13-21). These agents have shown accumula-
tion in several tumor types. Tl-201 is a potassium analog.
Its distribution depends on blood flow, BBB breakdown,
and metabolic activity with uptake by the Na+/K+ pump.
Tc-99m sestamibi is transported by the endothelial cell
to the mitochondrial activity. There is some accumula-
tion of Tc-99m sestamibi in the choroid plexus, which
may make it less than ideal in some tumors.
The procedure for SPECT tumor imaging involves
imaging approximately 20–30 minutes following injec-
tion of 2–4 mCi (74–148 MBq) of Tl-201 or 20 mCi (740
MBq) Tc-99m sestamibi. Occasionally, a 2-hour delayed
Tl-201 acquisition may be helpful as the abnormal tumor
tissue would be expected to washout more slowly than
normal brain or areas of BBB disruption. Visual analysis
typically shows uptake equal to or greater than the scalp
or the contralateral side in tumors. There is some over-
lap in the appearance of malignant and infectious
processes. An intracranial abscess, for example, often
has increased activity. Because infections usually have
Figure 13-21 Intracranial lymphoma. A mass is seen in a
patient with AIDS on the T2-weighted MRI (left; arrow) with the
Tl-201 SPECT showing intense tumor uptake (right).
lower uptake than malignancies, quantitative analysis
may help improve specificity. A region of interest ( ROI )
is drawn around the abnormal uptake and compared to
the contralateral normal. Delayed images may improve
sensitivity as the tumor retains activity and the back-
ground clears.
Movement Disorders
Movement disorders consist of a wide variety of dis-
eases. Some like Parkinson’s disease ( PD) have dimin-
ished movement states while others like chorea produce
excessive movement. PD is the most familiar disorder,
affecting approximately 1.5% of people over 65 and
2.5% of those over the age of 80.As the dopaminergic
neurons in the substantia nigra degenerate, patients
experience resting tremor, rigidity, and bradykinesia.
There are three groups of Parkinsonian syndromes: idio-
pathic PD, secondary PD ( due to disease such as
Wilson’s disease or other extrinsic agents such as carbon
monoxide poisoning and neuroleptic drugs), and neu-
rodegenerative syndromes such as multisystem atrophy
( MSA) and progressive supranuclear palsy ( PSP). Clinical
differentiation of idiopathic PD from etiologies such as
MSA may be difficult when patients do not respond to L-
dopa therapy.
The ability of PET to image the dopaminergic system
in patients with Parkinson’s disease and other move-
ment disorders has been known for decades. The cor-
pus striatum is the key to understanding PD in imaging
studies.The striatum consists of the lentiform nucleus
( putamen and globus pallidus) and caudate nucleus. As
the dopaminergic striatal neurons degenerate, effects
downstream in the globus pallidus can be imaged.
Diagnosis hinges on the ability to discriminate anterior
from posterior changes. Generally, it is not possible to
separate the components of the lentiform nucleus into
the putamen and globus pallidus by PET. Over the
 Central Nervous System 441

course of the disease, PD affects the posterior striatum

first and moves anteriorly: posterior putamen first, then
anterior putamen, then finally the caudate nucleus. It is
estimated that there is a 2–10% decrease in striatal activ-
ity per year in PD.
PET has largely been a research tool in movement
disorders, utilizing expensive agents such as ( F-18)
6-fluorodopa. Now, many new agents help study differ-
ent components of the dopamine neurotransmitter sys-
tem (Fig. 13-22). These radiopharmaceuticals are grouped
based on the location or mechanism of dopamine metab-
olism they image. The classic agent, F-18 dopa, enters the
dopamine metabolism pathway early as an analog of L-dopa
and measures dopamine neuron integrity and loss.
However, F-18 dopa tends to underestimate loss. Other
agents target the vesicular monoamine transporter type 2
( VMAT2 ), presynaptic membrane dopamine transporter
(DAT),and postsynaptic dopamine receptors (D2 and D1).
New tropane agents derived from cocaine have
been developed to image DAT activity. These include
F-18/C-11 β-CIT and SPECT agents I-123 β-CIT,Tc-99m
TRODAT-1, and I-123 IACFT ( ALTROPANE). These
agents may detect preclinical disease and differentiate
patients with Parkinson’s disease from those with non-
Parkinson’s tremor. Patients with Parkinson’s disease
show decreased radiotracer binding on the symptomatic
side, as well as often detecting preclinical disease on the
contralateral side ( Fig. 13-23).

Figure 13-23 Dopamine metabolism. A, Normal distribution

Huntington’s Chorea
The hereditary disorder Huntington’s chorea, also
called Huntington’s disease, usually manifests itself
between ages 35 and 50 years, and inevitably progresses
to uncontrollable choreiform movements and demen-
tia. The caudate and putamen are deficient in the
inhibitory neurotransmitter gamma-aminobutyric acid
of an experimental dopamine transporter agent, I-123 Altropane,
has high striatal uptake relative to cortex. B, Parkinson’s disease
demonstrates diffusely low striatal activity on the more severely
affected right side. Partial loss is seen on the left, sparing the left
caudate. This reflects the progression from unilateral to bilateral
involvement as well as from the posterior striatum forward.
(Courtesy of Dr. Robert Licho and Boston Life Sciences, Inc.)

( GABA) and in glutamic acid decarboxylase. Although

the disease can begin asymmetrically, symmetrical
involvement eventually develops. Both PET and SPECT
imaging can show decreased uptake in the caudate
nucleus, which often precedes the basal ganglia atrophy
seen on CT.

Figure 13-22 Dopamine neuron production and metabolism.
The sites of PET and SPECT agent uptake are shown. AAAD,
Aromatic amino acid decarboxylase; VMAT2, vesicular monoamine
transporter type 2; DAT, dopamine reuptake transporter.
Head Trauma
Tc-99m HMPAO SPECT is more sensitive than CT in
detecting abnormalities in patients with a history of
closed traumatic brain injury. SPECT can detect the
changes earlier, particularly in patients with minor head
injuries. In addition to evidence of acute injury in the
442 NUCLEAR MEDICINE: THE REQUISITES

form of coup and contrecoup rCBF deficits, SPECT stud- Choroid

ies can demonstrate residual flow defects in patients plexus
Superior sagittal sinus
with remote trauma, often in the temporal lobes. In one of lateral
ventricle Arachnoid
study, SPECT showed rCBF defects in 80% of patients
with head trauma versus 55% on CT and 45% on MRI.
With minor head injuries, 60% showed deficits on SPECT Choroid
and only 25% on CT. plexus
of third
ventricle
Psychiatric Diseases
Superior
The role of PET and SPECT imaging in psychiatric disease cistern
is uncertain. Although frontal lobe hypometabolism and
hypoperfusion have been described in schizophrenia, Confluence
the findings are nonspecific. Studies in patients with of the
Pontine cistern sinuses
depression have yielded conflicting results, although
patients with depression and metabolic disturbances Cerebellomedullaris
Choroid plexus cistern
usually have normal perfusion. Limited studies in of fourth ventricle
patients with obsessive-compulsive disorder, mostly with Foramen
PET, have shown increased metabolism in the orbital of Magendie
region of the frontal cortex and caudate nuclei.
Presently, functional scintigraphy may be of most value in
identifying underlying organic disease in patients with
psychiatric disease.

Figure 13-24 Flow dynamics of cerebral spinal fluid (CSF).

CISTERNOGRAPHY
Study of cerebrospinal fluid ( CSF) dynamics using radio-
tracers has been used for many years to diagnose a site of
CSF leakage, to determine shunt patency, and to manage
hydrocephalus. Although CT and MRI are now often
used, radionuclide cisternography can still play an impor-
tant role because of the unique physiologic information
it provides. To be effective, close coordination with
structural imaging studies and detailed knowledge of the
clinical problem are necessary.
An understanding of normal CSF dynamics is impor-
tant for image interpretation. CSF is secreted by ven-
tricular choroid plexus and, to a lesser extent, from
extraventricular sites. Normally, CSF drains from the lat-
eral ventricles through the interventricular foramen of
Monro into the third ventricle ( Fig. 13-24). With the
additional CSF produced by the choroid plexus of the
third ventricle, it then passes through the cerebral aque-
duct of Sylvius into the fourth ventricle and then leaves
the ventricular system through the median foramen of
Magendie and the two lateral foramina of Luschka. The
CSF then enters the subarachnoid space surrounding the
brain and spinal cord. Along the base of the brain,
the subarachnoid space expands into a number of lakes
called cisterns. The subarachnoid space extends over
the surface of the brain. The CSF is absorbed through
the pacchionian granulations of the pia arachnoid villi
into the superior sagittal sinus.
Originating in lateral ventricle choroid plexus, CSF flows
through the third and fourth ventricles into the basal cisterns,
moves over the convexities, and finally is reabsorbed in the
superior sagittal sinus.
Radiopharmaceuticals
Radiopharmaceuticals injected intrathecally into the
lumbar subarachnoid space must meet strict standards
for sterility and apyrogenicity. They should follow the
flow of the CSF without affecting the dynamics, and
they should rapidly clear through the arachnoid villi.
The chelating agent, diethylenetriamine pentaacetic
acid ( DTPA), is often used in renal imaging. DTPA is
ideal as it is nonlipid soluble, not metabolized, and not
absorbed across the ependyma before reaching the
arachnoid villi. Because imaging in cisternography
studies may extend over a period of days, indium-111
DTPA ( In-111 DTPA) is the agent of choice due to its
longer half-life ( 67 hours) and reasonable imaging
characteristics. Studies looking for CSF leaks may be
done with Tc-99m DTPA as its improved imaging char-
acteristics may improve sensitivity. CSF shunt patency
exams often utilize Tc-99m DTPA, although In-111
DTPA is used in situations requiring extended imaging.
Methods
Proper sterile lumbar puncture technique is critical and
should be done by an experienced clinician to ensure
 Central Nervous System 443

subarachnoid injection. An initial image of the injection

site ensures that the radiotracer was delivered to the cor- Table 13-3 Dosimetry for In-111 Cisternography
rect location. Early evaluation confirms dose migration
up the vertebral column and a lack of excessive renal Organ Rads/500 uCi (cGy/18.5MBq)
activity. Serial imaging is needed as described in the pro-
Total body 0.04
tocol in Box 13-8. Kidneys 0.22
Spinal cord
Surface 5.0
Pharmacokinetics Average 1.5
Radiotracer injected into the intrathecal space normally Brain
Surface 4.1
reaches the basal cisterns by 1 hour, the frontal poles and Average 0.50
sylvian fissure area by 2–6 hours, the cerebral convexities Bladder 0.50
by 12 hours, and the arachnoid villi in the sagittal sinus Testes 0.05
by 24 hours. Flow to the parasagittal region occurs Ovaries 0.06
through both central and superficial routes. The radio-
tracer does not normally enter the ventricular system
because physiological flow is in the opposite direction.

Clinical Applications
Dosimetry
Hydrocephalus
To some extent, the radiation absorbed dose depends on Hydrocephalus is abnormal enlargement of the CSF
the clearance dynamics of a particular patient. The spaces resulting from abnormalities of CSF production,
spinal cord receives the highest dose, followed by the circulation, or absorption ( Table 13-4). MRI and CT are
kidney and bladder, because the radiopharmaceutical most often used to select patients who might benefit
undergoes renal excretion ( Table 13-3). from intervention, whereas radionuclide cisternography
is generally reserved for situations that remain unclear.
When assessing hydrocephalus, it must first be known if
Box 13-8 Cisternography: Protocol the process is noncommunicating or communicating.
Summary Then the route of radiopharmaceutical administration
and expected pattern during cisternography can be pre-
PATIENT PREPARATION
dicted and evaluated.
In noncommunicating etiologies of hydrocephalus,
None.
there is an obstruction of flow from the ventricular sys-
RADIOPHARMACEUTICAL tem into the basal cisterns and subarachnoid space. This
Indium-111 DTPA, 250 μCi (9.3 MBq) is commonly due to a mass or congenital abnormality at
or above the fourth ventricle. The diagnosis is usually
INSTRUMENTATION made by MRI.
Camera: large-field-of-view gamma In communicating hydrocephalus, CSF is free to flow
Collimator: medium energy from the intraventricular region into the subarachnoid
IMAGING PROCEDURE
Inject slowly into lumbar subarachnoid space using a
22-gauge needle with the bevel positioned vertically.
Patient should remain recumbent for at least 1 hr after Table 13-4 Classification of Hydrocephalus
injection.
All images should be obtained for 50k counts. Classification Site of obstruction

IMAGING TIMES Obstructive

1 hr: thoracic-lumbar spine for evaluation of injection Noncommunicating Intraventricular between lateral
adequacy. ventricles and basal cistern
Communicating Extraventricular, affecting basal
3 hr: base of the skull to visualize basilar cisterns.
cisterns, cerebral convexities
24 and 48 hr: evaluation of ventricular reflux and
and arachnoid villi
arachnoid villi resorption. Nonobstructive
Obtain anterior, posterior, and both lateral views of the Generalized atrophy None
head at 3, 24, and 48 hr. Localized atrophy None
444 NUCLEAR MEDICINE: THE REQUISITES

space. The obstruction to CSF flow is extraventricular, in

the basal cisterns, cerebral convexities, or arachnoid villi. Table 13-5 Patterns of Cerebrospinal Fluid Flow
Common causes include previous subarachnoid hemor-
rhage, chronic subdural hematoma, leptomeningitis, and Cerebrospinal fluid
meningeal carcinomatosis all leading to poor CSF move- Type movement Etiologies
ment and reabsorption. On anatomical imaging, the ven-
I Basal cistern 2–4 hr Normal
tricular system is dilated out of proportion to the Sylvian fissure 6 hr Intraventricular obstructive
prominence of cortical sulci and the basal cisterns. It may Over convexities 24 hr hydrocephalus
be difficult to differentiate this extraventricular obstruc- Decreased activity 48 hr (noncommunicating)
tion from nonobstructive “hydrocephalus ex vacuo,” a sec- II No ventricular activity Cerebral atrophy
ondary expansion of the ventricles to fill the void Delayed migration over Advanced age
following neuronal tissue loss from atrophy or stroke. convexities
Most commonly, radionuclide studies help assess those IIIA Transient ventricular Intraventricular obstruction
communicating hydrocephalus patients with normal-pres- activity Cerebral atrophy
sure hydrocephalus ( NPH) to determine if the patient is Clearance by usual Evolving or resolving
likely to benefit from CSF shunting. NPH manifests clini- migration (often) communicating
hydrocephalus
cally with progressive dementia, ataxia, and incontinence.
Surgical shunting of CSF can potentially cure this cause of IIIB Transient ventricular Communicating
dementia, but not all patients improve with surgery. activity hydrocephalus with
Clearance but not by alternate reabsorption
Predicting which patients will respond is a diagnostic
usual migration pathway (trans-
problem. Radionuclide cisternography can be helpful to convexity ependymal)
when used in conjunction with clinical findings, such as
IV Persistent ventricular Communicating
a response (mental clearing) to CSF fluid reduction.
activity hydrocephalus
Inadequate clearance
Cisternography Image Interpretation
Several patterns of flow can be observed after introduc-
tion of radiopharmaceutical into the intrathecal space
been used to treat obstructive hydrocephalus. Com-
( Table 13-5). Normal flow should not reflux into the
plications may include catheter blockage, infection,
ventricles and should move over the convexities by
thromboembolism, subdural or epidural hematomas, dis-
24 hours ( Figs. 13-25 and 13-26).
connection of catheters, CSF pseudocyst, bowel obstruc-
In patients with noncommunicating hydrocephalus,
tion,and bowel perforation.
cisternography usually shows a normal pattern of flow
The diagnosis of shunt patency and adequacy of CSF
up to the basal cisterns, over the convexities. No ventric-
flow can often be made by examination of the patient
ular reflux is seen. However, if activity is injected into
and inspection of the subcutaneous CSF reservoir. When
the ventricles through a ventriculostomy rather than via
this assessment is uncertain, radionuclide studies with
lumbar puncture, serial images show minimal activity in
In-111 DTPA or Tc-99m DTPA are useful for confirming
the basal cisterns.
the diagnosis. Familiarity with the specific shunt type
In communicating hydrocephalus, including NPH
and its configuration is helpful. For example, the valves
patients, cisternography can show a spectrum of CSF
may allow bidirectional or only unidirectional flow.
flow patterns ( Fig. 13-27). The common denominator is
A proximal shunt limb consists of tubing running from
absent flow or a marked delay of activity flow up over the
convexities of the brain. Ventricular reflux of activity
may occur transiently or persist. Atrophy alone will
Type I Type II Type IIIA Type IIIB Type IV
cause delayed tracer movement through the enlarged sub-
arachnoid space, sometimes with transient ventricular
reflux. However, normal clearance over the hemispheres
is seen by 24 hours. It has been suggested that communi-
cating hydrocephalus patients with persistent ventricular
activity and no activity over the convexities ( the type IV
cisternographic pattern) are most likely to benefit from
shunting. Figure 13-25 Iodine-123 Isopropyl iodoamphetamine (IMP)
redistribution. Left, Immediate postinjection I-123 IMP transaxial
SPECT image shows a cortical perfusion defect in the left frontal
Surgical Shunt Patency cortex. Right, Repeat SPECT 4 hours later shows much improved
A variety of diversionary CSF shunts (ventriculoperitoneal, perfusion in a comparable transverse section, consistent with at
ventriculoatrial, ventriculopleural, lumboperitoneal) have least partially reversible ischemia.
Figure 13-26 Normal cisternogram. Anterior and lateral images 4 and 24 hours after intrathecal
radiotracer injection show normal transit up over the convexities with no ventricular reflux.

Figure 13-27 Communicating normal pressure hydrocephalus at 24 hours (top row), 48 hours
(middle row) and 72 hours (bottom row) in anterior (left), right lateral (middle) and left lateral (right)
projections. Ventricular reflux (closed arrowhead) is present,as is very delayed flow over the convexities
(open arrowhead). The intracerebral activity at 72 hours was caused by transependymal uptake.
446 NUCLEAR MEDICINE: THE REQUISITES

the ventricles into the reservoir while the distal limb car-
ries CSF away from the reservoir into the body. Box 13-9 Shunt Patency: Protocol
Shunt injection should be performed with aseptic Summary
technique by a physician familiar with the type of shunt
in place, preferably the neurosurgeon ( Box 13-9). PATIENT PREPARATION
Patency of the proximal shunt limb can sometimes be None.
evaluated before checking distal patency. In patients
with certain types of variable or low pressure two-way RADIOPHARMACEUTICAL
valves, the distal catheter is initially occluded by manu- Technetium-99m DTPA, 0.5 to 1 mCi (18.5–37 MBq), or
ally pressing on the neck. The pressure may cause indium-111 DTPA, 250 μCi (9.3 MBq)
injected tracer to flow into the proximal limb. INSTRUMENTATION
Images should show prompt flow into the ventricles
Camera: wide-field-of-view gamma.
followed by spontaneous distal flow through the shunt
Collimator: all purpose.
catheter ( Figs. 13-28 and 13-29). The shunt tubing is Computer and camera setup: 1-min images for 30 min.
usually seen. Catheters draining into the peritoneum
show accumulation of radiotracer freely within the IMAGING PROCEDURE
abdominal cavity. Using aseptic technique. Clean the shaved scalp with
Betadine.
Cerebrospinal Fluid Leak Penetrate the shunt reservoir with a 25- to 35-gauge
Trauma and surgery ( transsphenoidal and nasal) are needle.
the most common causes for CSF rhinorrhea. Once the needle is in place, position the patient’s head
Nontraumatic causes include hydrocephalus and con- under the camera with the reservoir in the middle
of the field of view.
genital defects. CSF rhinorrhea may occur at any site,
Inject the radiopharmaceutical.
from the frontal sinuses to the temporal bone ( Fig. 13-
Take serial images for 30 min.
30). The cribriform plate is most susceptible to fracture If no flow is seen, place the patient in an upright
which can result in rhinorrhea. Otorrhea is much less position and continue imaging for 10 min.
common. Accurate localization of CSF leaks can be clin- If still no flow is seen, obtain static images of 50k after
ically difficult. Although glucose oxidase test strips are 1 and 2 hr.
used to confirm CSF leak, both lacrimal and nasal secre- If flow is demonstrated at any point, obtain 50k images
tions contain glucose,and the false positive rate may be as of the shunt and tubing every 15 min until flow to the
high as 50%. distal tip of the shunt tubing is identified or for 2 hr,
Radionuclide studies are sensitive and accurate meth- whichever is first.
ods of CSF leak detection. To maximize the sensitivity To determine proximal patency of the reservoir, the
distal catheter can be manually occluded during the
of the test, nasal pledgets are placed in the anterior and
procedure so that the radiotracer will reflux into the
posterior portion of each nasal region by an otolaryngol-
ventricular system.
ogist and then removed and counted 4 hours later

Figure 13-28 Cerebrospinal shunt patency evaluation. A, Ventriculoperitoneal shunt at 10 min

(left) shows activity in the reservoir port and distal limb of the shunt moving down the neck and
chest. Intraventricular activity is also seen. By 30 minutes (middle), activity is in the abdomen with
free flow in the peritoneum (right).
Continued
 Central Nervous System 447

Figure 13-28, cont’d B, Ventriculopleural shunt with normal radiotracer flow through the
shunt into the pleural space which decreases over time.

Figure 13-29 Obstructed cerebrospinal shunt. After injection of Tc-99m DTPA into the
reservoir, good reflux into the ventricles is seen, consistent with patency of the proximal limb of the
shunt. However, no distal drainage occurs over 60 minutes from obstruction.
448 NUCLEAR MEDICINE: THE REQUISITES

CSF leak to:

Box 13-10 Cerebrospinal Fluid Leak
Frontal sinus Detection: Protocol Summary
Ethmoid sinus

Sphenoid sinus PATIENT PREPARATION

Nasal pledgets should be placed and labeled as to
location. The pledgets should be weighed before
placement.
After intrathecal injection, place patient in
Trendelenburg position to pool the radiotracer in the
basal regions until imaging begins.
Once radiotracer reaches basal cisterns, position patient
in a position that increases cerebrospinal fluid
leakage.
Figure 13-30 Common sites of cerebrospinal fluid leakage. Rhinorrhea: incline patient’s head forward and against
camera face with the camera positioned in the
lateral position.
Otorrhea: obtain posterior images instead of lateral
views.

RADIOPHARMACEUTICAL
In-111 DTPA, 250 μCi (9 MBq)

Sphenoid INSTRUMENTATION
sinus
Camera: large-field-of-view gamma
Collimator: medium energy

Olfactory
cleft
Middle
meatus
Spheno-
ethmoid
recess
Figure 13-31 Placement of pledgets for cerebrospinal fluid
leak study. The labeled cotton pledgets are placed by an otolaryn-
gologist at various levels within the anterior and posterior nares to
detect leakage from the frontal, ethmoidal, and sphenoidal sinuses.
( Fig. 13-31). A ratio of nasal-to-plasma radioactivity
greater than 2:1 or 3:1 is considered positive. The radio-
tracer is injected intrathecally via aseptic lumbar punc-
ture ( Box 13-10).
The site is most likely to be identified during a time
when heavy leakage is occurring. Often, the patient
position associated with greatest leakage is reproduced
during imaging. Imaging in the appropriate projection is
important for identifying the site of leak; lateral and ante-
rior imaging is used for rhinorrhea and posterior imaging
for otorrhea.
Scintigraphic studies show CSF leaks as an increasing
accumulation of activity at the leak site (Fig. 13-32).
IMAGING PROCEDURE
Setup
Inject intrathecally 500 μCi (18 MBq) of In-111 DTPA in
5 ml of dextrose 10% in water.
Begin imaging when activity reaches the basal cisterns
(1 to 4 hr).
ACQUISITION
Acquire 5 min/frame for 1 hr in the selected view, then
acquire anterior, left lateral, right lateral, and posterior
views.
Obtain 50k images every 10 min for 1 hr in the original
view.
Remove pledgets and place in separate tubes. Draw a
5-ml blood sample.
Count pledgets and 0.5-ml aliquots of plasma.
Repeat views may be indicated at 6 and 24 hr.
Calculate the ratio of pledgets-to-plasma activity:
pledget counts/pledget capacity divided by serum
counts/0.5 ml.
INTERPRETATION
Positive for CSF leakage if the pledget/plasma activity
ratio is greater than 2–3:1.
However, counting the pledgets is more sensitive than
imaging for detecting CSF leaks. Pledgets are also help-
ful in determining the origin of the leak (anterior versus
posterior).
 Central Nervous System 449

Figure 13-32 Positive radionuclide CSF leak study. In-111 DTPA left lateral views show
increasing radioactivity over time originating from the nares and leaking into the nose and
mouth (arrowheads).

SUGGESTED READING Mountz JM, Liu HG, Deutsch G: Neuroimaging in cerebrovascu-

Carmago EE: Brain SPECT in neurology and psychiatry. J Nucl
Med 42:611, 2001.
Herholz K, Herscovitch P, Heiss WD: NeuroPET. Berlin,
Springer, 2004.
lar disorders: Measurement of cerebral physiology after stroke
and assessment of stroke recovery. Sem Nucl Med 33:56, 2003.
Van Heertum RL,Tikofsky RS: Cerebral SPECT Imaging, ed. 2.
New York, Raven Press, 1995.
 14
CHAPTER Cardiac System

Myocardial Perfusion Scintigraphy Data Analysis and Study Interpretation

Radiopharmaceuticals Qualitative Analysis
Thallium-201 Chloride Quantitative Data Analysis
Tc-99m Sestamibi (Cardiolite) Clinical Applications
Tc-99m Tetrofosmin (Myoview) Coronary Artery Disease
Other Single-Photon Perfusion Agents Valvular Heart Disease
Scintigraphic Methodology Cardiomyopathy and Myocarditis
Planar Imaging Assessment of Cardiac Toxicity
Single-Photon Emission Computed Tomography (SPECT) Pulmonary Disease
Gated SPECT Congenital Heart Disease
Diagnosis and Evaluation of Coronary Artery Disease Infarct-Avid Imaging
Physiology of Ischemia Tc-99m Pyrophosphate
Cardiac Stress Testing Mechanism of Localization
Stress Perfusion Scintigraphy Methodology
Exercise Stress Testing Scintigraphic Patterns in Acute Myocardial Infarction
Pharmacologic Stress Testing Clinical Applications and Utility
One- and Two-Day Protocols Investigational Cardiac Radiopharmaceuticals
Image Interpretation I-123 Meta-Iodobenzyl-Guanidine
Quantitative Analysis Fatty Acid Radiopharmaceuticals
Sensitivity and Specificity
Prognosis and Risk Stratification Radionuclides were first used in 1927 by Blumgart to
Positron Emission Tomography of the Heart investigate the circulatory system. In 1948, Prinzmetal
Radiopharmaceuticals for Myocardial Perfusion performed the first cardiac study by injecting a subject
Nitrogen-13 Ammonia with radiolabeled albumin and, with the use of a Geiger
Rubidium-82 Chloride counter placed on the chest, recorded a time–volume
Oxygen-15 Water curve. Modern-day clinical cardiac radionuclide
Diagnosis of Coronary Artery Disease imaging techniques were developed in the 1970s and
PET Metabolic Radiopharmaceutical 1980s. Of these, some have become routine studies
Fluorine-18 Fluorodeoxyglucose used today, including myocardial perfusion scintigra-
Detection of Myocardial Viability phy and radionuclide ventriculography ( RVG). Over
Combined Tc-99m Sestamibi and F-18 FDG Imaging the years, there have been marked advancements in
Radionuclide Ventriculography instrumentation, methodologies, processing, display,
Radiopharmaceuticals and quantification.
Blood Pool Agents Numerous other diagnostic studies are available to the
First-Pass Agents cardiologist,such as electrocardiography ( ECG),echocar-
Acquisition Techniques diography, magnetic resonance imaging ( MRI), contrast
First-Pass Studies angiography, and computed tomography ( CT) angiogra-
Equilibrium Gated Blood Pool Studies phy. The continuing value of cardiac nuclear studies

450
 Cardiac System 451

stems from their noninvasiveness and accurate portrayal Thallium-201 Chloride

of a wide range of functional and metabolic that predict Chemistry
prognosis and risk. Thallium is a metallic element in the IIIA series of the
Periodic Table (see Figure 1-1). In pharmacological
doses, thallium is a poison, but it is nontoxic when used
in subpharmacologic tracer doses. It is administered to
MYOCARDIAL PERFUSION the patient in the chemical form of thallium chloride.
SCINTIGRAPHY Physics
The radionuclide Tl-201 is cyclotron produced. It decays
Myocardial perfusion scintigraphy is currently by far the by electron capture to its stable mercury-201 daughter
most commonly performed cardiac nuclear study, with a physical half-life of 73 hours. The photons avail-
constituting approximately one third of all nuclear med- able for imaging are mercury K-characteristic x-rays in
icine procedures done annually in the United States, the range of 69 to 83 keV (95% abundant) and gamma
with substantial growth annually. Gated myocardial rays of 167 keV (10%) and 135 keV (3%) ( Box 14-1). For
perfusion scintigraphy plays an important role in the clinical gamma-camera imaging, a 20-30% window is cen-
diagnosis, prognosis, risk assessment, and management tered at 69–83 keV and a 20% window at 167 keV.
of heart disease. Mechanism of Localization and Pharmacokinetics
Myocardial perfusion scintigraphy depicts two sequen- After intravenous injection, blood clearance is rapid
tial physiological events. First, the radiopharmaceutical ( Fig. 14-1). Tl-201 is transported across the myocardial
must be delivered to the myocardium. Second, a viable cell membrane via the Na+–K+ ATPase pump. Greater
metabolically active myocardial cell must be present to than 85% of Tl-201 is extracted by the myocardial cell
extract the radiotracer. The scintigraphic images are on first pass through the coronary capillary circula-
a map of regional myocardial perfusion to viable tion ( Table 14-1). Peak myocardial uptake occurs by
myocardial tissue. If a patient has a decrease in regional 10 minutes. Approximately 3% of the administered dose
perfusion due to hemodynamically significant coronary localizes in the myocardium. The biological half-life is
artery disease or a loss of cell viability as a result of approximately 10 days.
myocardial infarction, a photon-deficient or cold region Extraction is proportional to relative regional perfu-
is seen on the perfusion images. All diagnostic patterns sion over a wide range of flow rates ( Fig. 14-2). At very
in the many diverse applications follow from these high flow rates, extraction efficiency decreases; at very
observations. low rates, it increases. Tl-201 can only be extracted by
viable myocardium and does not concentrate in regions
of infarction or scarred myocardium.
Radiopharmaceuticals
Radiolabeled potassium ( K+) was first considered for
myocardial perfusion imaging because it is the major Box 14-1 Physical Characteristics of
intracellular cation. Sodium ( Na+)–K + homeostasis is Thallium-201 and Technetium-
maintained as an energy-dependent process involving 99m (Sestamibi and
the Na+–K+ ATPase ( adenosine triphosphatase) pump in Tetrofosmin)
the myocardial cell membrane. However, neither K +
nor its analogs, cesium and rubidium, were found suit-
Tc-99m
able for single-photon imaging because of their high-
(Sestamibi
energy photons. Rubidium ( Rb-82) has found use in Physical and Tetro-
dual-photon imaging and positron emission tomogra- Characteristic Thallium-201 fosmin)
phy ( PET).
The first radiopharmaceutical used on a clinical basis Mode of decay Electron capture Isomeric
for perfusion imaging was thallium-201 chloride (Tl-201). transition
Tl-201 behaves physiologically much like K+, although it Physical half-life 73 hours 6 hours
is not a true K+ analog in a chemical sense. First used for Principal Mercury x-rays Gamma rays
emissions 69–83 keV 140 keV (89%)
myocardial scintigraphy in the mid-1970s,it remained the
(abundance)* Gamma rays
only perfusion agent available until 1990. Two Tc-99m-
167 keV (10%)
labeled cardiac perfusion radiopharmaceuticals, ses- 135 keV (2.7%)
tamibi and tetrofosmin, were introduced. All three are
now routinely used clinically for myocardial perfusion *Abundance is the percentage of time each emission type that occurs
scintigraphy. with each decay.
452 NUCLEAR MEDICINE: THE REQUISITES

Scintigraphy obtained early after initial uptake reflects

capillary blood flow. After uptake, Tl-201 undergoes
redistribution,which is a continual dynamic exchange of
Tl-201 between the myocardial cell and the vascular
blood pool ( Fig. 14-1). As Tl-201 leaves the myocardial
cell, it is replaced by Tl-201 circulating in the systemic
blood pool, which is also undergoing Tl-201 redistribu-
tion. Several hours after injection, the images depict an
equilibrated pattern reflecting regional blood volume or
myocardial K+ space. This redistribution occurs after

Figure 14-1 Thallium-201 pharmacokinetics: redistribution.

After intravenous injection,Tl-201 clears rapidly from the blood
pool. Normal stress peak myocardial uptake occurs by 10 minutes.
Redistribution begins promptly after initial uptake.This is a
constant dynamic exchange of thallium between the myocytes and
blood pool. Normal myocardium progressively clears over 3 hours.
In the presence of ischemia, uptake is delayed and reduced and
clearance is slow.With infarction, there is little uptake and very
little change over time.The schematic diagram relates thallium
pharmacokinetics to scintigraphic findings.The ischemic region,
although initially hypoperfused compared to the normal region Figure 14-2 Myocardial perfusion radiopharmaceuticals:
equalizes scintigraphically at 3 hours. Modified with permission uptake relative to coronary blood flow.An ideal myocardial
from Dilsizian V, Narula J: Atlas of Nuclear Cardiology. perfusion tracer would show a linear relationship to blood flow
Philadelphia, Current Medicine, 2003. over a wide range of flow rates.At rest, the normal myocardial
coronary low rate is 1 mL/g/min.With exercise, it can increase to
2 mL/g/min.With pharmacologic vasodilation, flow may exceed
2 mL/g/min.Tl-201,Tc-99m sestamibi and Tc-99m tetrofosmin all
have extraction that is proportional to blood flow but
underestimate flow at high flow rates.

Table 14-1 Pharmacokinetics of Thallium-201, Tc-99m Sestamibi, and Tc-99m Tetrofosmin

Thallium-201 Tc-99m sestamibi Tc-99m tetrofosmin

Chemical class/charge Element cation Isonitrile cation Diphosphine cation

Mechanism of uptake Active transport Passive diffusion Passive diffusion
Na-K ATPase Negative Negative
pump electrical potential electrical potential
Myocyte localization Cytosol Mitochondria Mitochondria
Intracellular state Free Bound Bound
Preparation Cyclotron Generator/kit Generator/kit
First pass extraction fraction 85% 60% 50%
Percent cardiac uptake 3% 1.5 % 1.2%
Myocardial clearance 4 hr T1/2 Minimal Minimal
Body clearance Renal Hepatic Hepatic
Imaging time after injection
Stress 10 min 15–30 min 5–15 min
Rest 3–4 hrs 30–90 min 30 min
 Cardiac System 453

cardiac stress but also during a rest study. With normal
perfusion, the initial capillary blood flow image and the
delayed regional blood volume image are similar.
The unique pharmacokinetics of Tl-201 are the basis
for the “stress-redistribution” Tl-201 imaging strategy
that has been used in the detection of coronary artery
disease. Regions of decreased perfusion (cold, photon-
deficient) seen on early images after stress can be due
to either decreased blood flow (ischemia) or to the lack
of viable cells to fix the tracer (infarct). If an initial per-
fusion defect persists on delayed images, it depicts
infarction. Those defects showing “fill-in” of Tl-201
between stress and rest are viable myocardium ren-
dered ischemic during stress ( Fig. 14-1).
Dosimetry
Tl-201 has relatively high-radiation dosimetry because of
its long biological and physical half-life, thus limiting the
maximum allowable administered dose to 3.5–4.0 mCi
(130–150 MBq). The kidney receives the highest radia-
tion dose ( critical target organ), 5.1 rads/3 mCi
(5.3 cGy/111 MBq) ( Table 14-2).
Myocardial image quality is poor compared to Tc-99m
agents for several reasons. The low administered dose
results in a poor count rate. Thallium’s relatively low
photoenergy emissions (69–83 keV) compared to Tc-99m
( 140 keV) makes it poorly suited for modern-day gamma
cameras. The lack of a distinct photopeak and its high
scatter fraction further contributes to suboptimal image
quality. The low count rate makes gated SPECT problem-
atic and first-pass studies impractical.
Tc-99m Sestamibi (Cardiolite)
Tc-99m labeled sestamibi was approved by the Food and
Drug Administration ( FDA) for clinical use in 1990.
Chemistry
Sestamibi is a lipophilic cation and member of the chem-
ical isonitrile family. Tc-99m is surrounded by six isoni-
trile ligands (chemical name: hexakis 2-methoxyisobutyl
isonitrile). The radiopharmaceutical is prepared from
a manufacturer-provided kit.
Mechanism of Localization and Uptake
Being lipid soluble,Tc-99m sestamibi diffuses from the
blood into the myocardial cell and is retained in the
region of mitochondria because of its negative trans-
membrane potential. The first-pass extraction fraction
is lower than that of Tl-201, approximately 60% ( Table
14-1). As with Tl-201, extraction is proportional to
coronary flow and is underestimated at high flow rates
and overestimated at low flow ( Fig. 14-2).
Pharmacokinetics
After clearing rapidly from the blood, myocardial uptake
is prompt, although initially obscured by high uptake in
the lung and liver. Unlike the redistribution of Tl-201,
the clearance half-time from the myocardium is long,
and for practical purposes, this tracer remains fixed
within the myocardium. This allows for an imaging
time window of several hours after administration.
Because of renal and biliary excretion, there is progres-
sive clearance of liver and lung activity, thus improving
the myocardium-to-background activity ratios over time.
Imaging is usually begun 45–60 minutes after tracer
administration for resting studies and at 30 minutes for
exercise stress studies, in contrast to Tl-201 where imag-
ing must be started at 10 minutes before redistribution
has occurred.
Dosimetry
The radiation absorbed dose to the patient is relatively
low because of the Tc-99m label. The colon is the criti-
cal target organ, receiving about 5.4 rads/30 mCi (5.4
cGy/1110 MBq) ( Table 14-2).

Table 14-2 Radiation Dosimetry for Tl-201, Tc-99m Technetium-99m Tetrofosmin (Myoview)
Sestamibi, Tc-99m Tetrofosmin
Chemistry
This Tc-99m-labeled myocardial perfusion agent was
Tc-99m Tc-99m approved by the FDA for clinical use after sestamibi in
Tl-201+ sestamibi tetrofosmin
1996. It is a member of the diphosphine chemical class
Dose cGy/111 MBq cGy/1110 MBq cGy/1110 MBq [chemical name: 6,9-bis ( 2-ethoxyethyl)-3,12-dioxa-
(rads/3 mCi) (rads/30 mCi) (rads/30 mCi) 6,9 diphospha-tetradecane]. It is prepared from a kit.
Heart wall 0.3 0.5 0.5 Mechanism of Localization and Uptake
Liver 1.1 0.8 0.5 Similar to sestamibi,Tc-99m tetrofosmin is a lipophilic
Kidneys 5.1 2.0 1.4
Gallbladder 0.9 2.8 5.4
cation that localizes and is retained within mitochondria.
Urinary 0.6 2.0 2.1 Tc-99m tetrofosmin remains fixed in the heart, also simi-
bladder lar to sestamibi.
Colon 0.8 5.4 3.4 Pharmacokinetics
Thyroid 3.0 0.7 0.6 Tc-99m tetrofosmin is cleared rapidly from the blood.
Testes 0.9 0.3 0.4
Ovaries 1.1 1.5 1.1
Myocardial uptake is prompt. First-pass extraction is
Total body 0.4 0.5 0.2 slightly less than sestamibi (50%) with roughly 1.2% of the
injected dose taken up in the myocardium by 5 minutes
Target organ (highest radiation absorbed dose) in boldface type. after injection (Table 14-1). Extraction is proportional to
454 NUCLEAR MEDICINE: THE REQUISITES
blood flow, but underestimated at high flow rates, similar
to Tc-99m sestamibi (Fig. 14-2).
Heart-to-lung and heart-to-liver ratios improve with
time because of physiological clearance through the liver
and kidneys. Heart-to-liver ratios are somewhat higher
for Tc-99m tetrofosmin than sestamibi,allowing for earlier
imaging. After stress, imaging at 15 minutes is feasible.
Rest studies can be started 30 minutes after injection.
Dosimetry
The radiation absorbed dose to the patient is similar to
Tc-99m sestamibi ( Table 14-2), although the package
insert states that the gallbladder receives the highest
dose, 5.4 rads/20 mCi (5.4 cGy/1110 MBq). The reason
for the discrepancy is probably because of the variability
of gallbladder filling and emptying, depending primarily
on when the patient eats. Otherwise the colon would
receive the highest absorbed dose.
Other Single Photon Perfusion Agents
Tc-99m teboroxime (CardioTec,Bracco) is a drug approved
by the FDA in 1990, but to date has not found general clini-
cal use, although its proponents feel that its unique charac-
teristics will eventually make it a clinical agent. This neutral
lipophilic radiopharmaceutical comes from a class of com-
pounds referred to as boronic acid adducts of technetium
dioxime ( BATO). The extraction fraction is greater than
Tl-201. Uptake is proportional to flow but decreases with
increasing flow. Blood clearance half-time is less than
1 minute.
Myocardial clearance is extremely rapid (half-time of
5–10 minutes). Washout is proportional to regional
blood flow. Redistribution does not occur. The rapid
uptake and clearance dictate a very narrow window for
imaging, between 2 and 6 minutes resulting in relatively
poor-quality SPECT images because of low counting sta-
tistics and rapidly changing distribution.
Anterior
wall
LAD
Septum
LAD
LAD
and/or
RCA
RCA
Apex
Inferior
wall
Figure 14-3 Planar scintigraphy: relationship of coronary artery vascular supply to ventricular
wall segments.Anterior, left anterior oblique, and left lateral projections. LAD, Left anterior
descending artery; LCx, left circumflex branch; RCA, right coronary artery.
Technetium-99mN-NOET ( bis [N-etoxy, N-ethyl dithio-
carbamato] nitride technetium [V] is an investigational
myocardial perfusion radiopharmaceutical. It is a neutral
lipophilic compound with a first-pass extraction of 75%
at rest and 85% under hyperemic conditions. After pas-
sive diffusion, uptake occurs through linkage to proteins
bound in the lipid membrane of myocytes. Although
clearance from the myocardium is slow, it significantly
redistributes over time, predominantly by differential
washout. Its clinical role has not been determined.
Scintigraphic Methodology
Planar Imaging
Two-dimensional planar imaging was the standard imag-
ing methodology used for many years. Its accuracy for
the diagnosis of coronary artery disease was generally
good. However, accurate localization of regional perfu-
sion abnormalities predicts with only moderate success
the coronary artery bed involved ( Fig. 14-3).
Interpretation of planar images is limited by the high
background,overlapping structures,and the standard three
views acquired (left anterior oblique [LAO], right anterior
oblique [RAO], left lateral) ( Fig. 14-4). More accurate
assessment of regional perfusion became increasingly
important as myocardial perfusion scintigraphy was less
frequently requested for diagnosis and increasingly utilized
for prognosis,risk stratification,and patient management.
In current practice, planar imaging is limited to
patients who are severely claustrophobic who cannot
tolerate SPECT or for those who exceed weight restric-
tions on the SPECT imaging table. The anterior, left ante-
rior oblique, and left lateral ( or posterior oblique)
projections are routinely acquired ( Boxes 14-2 and 14-3).
Acquisition of the stress and rest images in the same pro-
jection is critical for proper interpretation ( Fig. 14-5).
Anterior
wall
LAD
Lateral
wall LAD
LCx and/or
RCA
LAD Apex
and/or RCA
RCA
Inferior
Apex
wall
 Cardiac System 455

Figure 14-4 Comparison of stress Tl-201 (A) and Tc-99m sestamibi (B) planar scintigraphy in the
same patient.Although the myocardium is well-visualized with both agents, the count rate with
Tl-201 is less than with Tc-99m sestamibi and the target-to-background ratio is lower, accounting for
its poor image quality.An attenuation artifact caused by interposition of subdiaphragmatic structures
can be seen on the lateral view; the effect is greater on the Tl-201 study.

Single-Photon Emission Computed Tomography
( SPECT)
SPECT has become the standard methodology for
myocardial perfusion scintigraphy. The cross-sectional
images have high-contrast resolution and are displayed
three-dimensionally ( Figs. 14-6 and 14-7), providing good
delineation of the various regional myocardial perfusion
beds supplied by their individual coronary arteries
( Fig. 14-8).
SPECT instrumentation and methodology are dis-
cussed in some detail in Chapter 4. The acquisition and
processing parameters are dictated to some extent by
the specific SPECT camera and computer software
employed; however, there is room for individual prefer-
ence. Variations include the number and position of
camera detector heads ( Fig. 14-9), the choice of acquisi-
tion method (e.g., continuous versus step and shoot),
number of stops, acquisition time for each angle, and
shape of orbit (e.g., elliptical or body contour versus cir-
cular). Elliptical orbits, although desirable because the
camera is closer to the body, often result in unacceptable
artifacts. Circular orbits are usually obtained.
Patient movement is a source of image degradation
with SPECT and dictates imaging time to be as short as
possible while acquiring sufficient counts for good image
quality. Image acquisition usually lasts 25–35 minutes.
SPECT can be obtained using a rotating single-headed
gamma camera; however, multiple detectors are advanta-
geous because a higher number of counts can be acquired
in a shorter period of time. Three-headed cameras pro-
vide the highest count rate; however, two-headed SPECT
cameras are the rule because of their versatility. They can
be used for various other types of scintigraphic studies in
the clinic.
456 NUCLEAR MEDICINE: THE REQUISITES
Box 14-2 Protocol: Thallium-201
Myocardial Perfusion Imaging
PATIENT PREPARATION AND FOLLOW-UP
Patients should fast for 4 hr prior to study
EKG leads should be moved out of field of view
DOSAGE AND ROUTE OF ADMINISTRATION
3 to 3.5 mCi (111 to 120 MBq) thallium-201 chloride
intravenously
TIME OF IMAGING
10 min after radiopharmaceutical administration
SPECT IMAGING ACQUISITION PARAMETERS
Collimator: low energy general-purpose parallel hole
Photopeak: 20% window centered at 80 and 167 keV
Patient position: supine, left arm raised
Rotation orbit: circular or elliptical
Matrix: 64 × 64 word mode
Arc and framing: 64 views, 180º (45º RAO, 135º LPO),
20 seconds per view
SPECT RECONSTRUCTION PARAMETERS*
Filter: Butterworth; cutoff 0.5 and order 8
Attenuation correction: review with and without
correction
Reconstruction technique: filtered backprojection or
iterative reconstruction
Image format: transaxial short axis, horizontal and
vertical long axis
PLANAR IMAGING
Collimator: low-energy, general purpose, parallel hole
collimator
Photopeak: 20% window centered at 80 and 167 keV
Image acquisition: anterior, 45º left anterior oblique
(LAO), and left lateral for 10 min each
Acquire rest and stress images in identical projections
*Choice of SPECT acquisition and reconstruction parameters is
influenced by the equipment used.
Because the heart lies in the anterior lateral chest
and considerable cardiac attenuation results in the pos-
terior projections, SPECT is often acquired over a 180-
degree arc from the left posterior oblique ( LPO) to the
right anterior oblique ( RAO) projection. The left arm is
positioned above the head to prevent attenuation.
A two-headed camera with the detectors at 90 degrees
takes maximum advantage of the 180-degree acquisi-
tion ( Fig. 14-9). The higher count rate of the multiple
Box 14-3 Protocol: Tc-99m Sestamibi and
Tetrofosmin Myocardial
Perfusion Imaging
PATIENT PREPARATION
Fasting for 4 hours
DOSE AND ROUTE OF RADIOPHARMACEUTICAL
ADMINISTRATION
10 to 30 mCi (370 to 1110 MBq) intravenously, see
individual protocols below
SPECT IMAGING PROTOCOL
1-Day rest/stress imaging
Rest: 370 MBq (10 mCi); imaging at 30 to 90 min
Stress: 1110 MBq (30 mCi); imaging at 15 to 30 min
2-Day rest/stress or stress/rest imaging: 1110 MBq
(30 mCi)
SPECT ACQUISITION PARAMETERS
Patient position: supine, left arm raised (180º arc)
Rotation: counterclockwise
Matrix: 128 × 128 word mode
Image/arc: 64 views (180º, 45º right anterior oblique,
135º left posterior oblique)
SPECT RECONSTRUCTION PARAMETERS*
Ramp filter
Convolution filter: Butterworth
Attenuation correction: review images with and without
correction
Oblique angle reformatting: short axis, vertical long axis
and horizontal long axis
Gated SPECT
ECG synchronized data collection: R wave trigger
8 Frames/cardiac cycle
PLANAR IMAGING PROTOCOL
Collimator: High-resolution
Window: 20% centered at 140 keV
For single-day rest and stress studies give 370 MBq
(10 mCi) at rest and image at 30 to 60 min
Rest studies: Begin imaging at 60 to 90 min after tracer
injection
Obtain anterior, 45 left anterior oblique (LAO), and left
lateral images
Obtain 750,000 to 1 million counts per view
Wait 4 hours and give 1110 MBq (30 mCi) with repeat
imaging at 15 to 30 min
Stress studies: Begin imaging at 15 to 30 min after tracer
injection
Obtain stress and rest images in identical projection
*Choice of SPECT acquisition and reconstruction parameters is highly
influenced by the equipment used. Protocols should be established in
each nuclear medicine unit for available cameras and computers.
 Cardiac System 457

Figure 14-5 Planar stress and rest Tc-99m sestamibi scintigraphy.The rest images are
considerably noisier because of the lower administered dose (8 mCi) compared to the stress images
(25 mCi).The images should be acquired so that the three planar views (anterior, left anterior
oblique, left lateral) are identical for optimal comparison.This study was interpreted as normal.

detectors makes possible the use of high-resolution col-
limators.
Filtered backprojection has been the standard method
used for cross-sectional image reconstruction; however,
with faster computers, iterative reconstruction tech-
niques are increasingly available and used. Software fil-
ters are chosen to optimize image quality by optimizing
the trade-off between high-frequency noise and lower
frequency oversmoothing.
Unlike other SPECT displays with axial transverse,
sagittal, and coronal slices, the heart is cut along its long
and short axes ( Figs. 14-10 and 14-11). This has
become the standardized method for CT, MRI, and
echocardiography.These SPECT cross-sectional images
more clearly depict the regional perfusion of the
myocardium as it relates to the coronary artery supply-
ing blood to that region ( Box 14-4) and permits visual
estimation of the degree and extent of the perfusion
abnormality ( Fig. 14-12). Various different qualitative
and quantitative methods have been used.
Gated SPECT
The high count rate available from 20–30 mCi (740–1110
MBq) of Tc-99m sestamibi or tetrofosmin and multidetector
systems makes it a feasible and common practice to per-
form ECG gating. Gated SPECT adds a cinematic three-
dimensional display of the contracting myocardium. Data
collection is triggered from the R-wave of the ECG and
arrhythmic beats are filtered out of the data collection
cycle. The cardiac cycle is typically divided into eight
frames,less than the 16 usually used for planar radionuclide
Tc-99m labeled erythrocyte ventriculograms, because of
limitations related to computer memory. The eight frames
limit, to a mild degree, the temporal resolution (pinpoint-
ing end-diastole and end-systole) and thus the accuracy of
ejection fraction calculation. Edge-detection software pro-
grams automatically draw endocardial and epicardial
regions of interest for ejection fraction calculation and wall
thickening analysis.
Diagnosis and Evaluation of Coronary
Artery Disease
A recurrent theme in nuclear medicine is that of enhanc-
ing diagnostic information by applying an interventional
maneuver to alter organ function, often while testing
functional reserve. Stress cardiac myocardial perfusion
scintigraphy and radionuclide ventriculography are ele-
gant examples of this principle.
458 NUCLEAR MEDICINE: THE REQUISITES

L
P
1 2 3 4 9 10 11 12

B
5 6 7 8 13 14 15 A
S
1
E

HSA HSA
ANT A
S L

INF I
1 2 3 4 N 1 2 3 4
F

1
L A
5 6 7 A 5 6 7 N
T 1 T
1

Figure 14-6 Normal SPECT stress thallium-201. Short-axis (top four rows), vertical long-axis
(bottom left), and horizontal long-axis views (bottom right).The top row for each slice orientation
is the immediate poststress study and the bottom row shows the 3-hour delayed images.

Physiology of Ischemia Coronary angiographic laboratories generally con-

Under resting conditions, coronary artery stenoses of up to
90% are not usually associated with a perfusion abnormality
(Fig. 14-13). The myocardial oxygen demand is low and
blood flow adequate. By increasing cardiac work, exercise
stress increases the demand for oxygen and increased blood
flow. Maximum exercise can increase coronary flow by
three to five times by coronary dilation.
Coronary flow reserve across a fixed mechanical
stenosis is limited. If exercise is vigorous, myocardium
in the watershed of a coronary artery with a hemo-
dynamically significant stenosis becomes ischemic.
Regionally reduced blood flow results in less delivery
and localization of the myocardial perfusion radiophar-
maceutical. This is seen on scintigraphic images as a rel-
atively cold defect in the ischemic region surrounded by
normal blood flow in the adjacent normal regions of the
heart ( Fig. 14-A1 [see color insert]).
sider stenoses of greater than 70% to be clinically signifi-
cant based on the rapid falloff in flow reserve
augmentation above this level ( Fig. 14-13). However, the
anatomical degree of stenosis has a poor correlation with
flow reserve and the degree of ischemia. Other factors
affect the functional significance of an anatomical cir-
cumferential narrowing ( e.g., the length, shape, and loca-
tion of a stenotic lesion). Thus functional imaging of
myocardial perfusion scintigraphy is often needed to
evaluate the clinical significance of a known stenosis,
particularly in the range of 50–70%.
Cardiac Stress Testing
Cardiac stress testing with ECG monitoring is an intervention
that has been used for years by the cardiologist to diagnose
coronary artery disease. A prerequisite of intervention is that
the degree of stress must be sufficient to unmask underlying

Figure 14-7 Normal SPECT stress and rest Tc-99m sestamibi. Note the excellent visualization of
the left ventricular myocardium.The more proximal short-axis views (top rows, far right)
demonstrate decreased uptake in the region of the membranous septum.
abnormalities ( Box 14-5). Graded treadmill exercise is
the standard method for cardiac stress testing. Exercise
increases cardiac work load and oxygen demand. The
treadmill study allows for assessment of the patient’s
functional cardiac status by directly monitoring exercise
tolerance, heart rate, blood pressure, and ECG response
to graded exercise. Indications and contraindications for
cardiac stress testing are listed in Box 14-6.
Exercise-induced myocardial ischemia produces
characteristic ST-T segment depression on ECG caused
by alterations in sodium and potassium electrolyte flux
across the ischemic cell membrane. The adequacy of
exercise is judged by the degree of cardiac work. Heart
rate and blood pressure provide such an indication.
Patients achieving >85% of the age-predicted maximum
heart rate ( 220 − age = maximum predicted heart rate)
are considered to have achieved adequate exercise
stress. The heart-rate/blood-pressure product, METS
(metabolic equivalents), and length of exercise (min-
utes) are other indicators used to judge the adequacy of
exercise. Failure to achieve adequate exercise is the
most common reason for a false negative stress test
( Box 14-7).
Most cardiologists perform cardiac treadmill exercise
in their offices. Although it provides clinically useful
diagnostic and functional information necessary to man-
Cardiac System 459
age patients, the overall accuracy of the cardiac treadmill
exercise test for the diagnosis of coronary artery disease
is modest, in the range of 75%, with numerous false nega-
tives and false positives. Specificity is particularly poor
in women, in patients with resting ECG ST-T changes, left
ventricular hypertrophy, bundle branch block, and
patients on digoxin. These patients often require myo-
cardial perfusion scintigraphy to confirm or exclude the
diagnosis of coronary artery disease.
Stress Perfusion Scintigraphy
The combination of stress testing with myocardial per-
fusion imaging is a commonly performed nuclear medi-
cine study. The overall accuracy for the diagnosis of
coronary artery disease is considerably higher for stress
myocardial perfusion imaging than the exercise tread-
mill study. SPECT provides valuable information on the
degree and severity of coronary artery disease that
allows for risk assessment, prognosis, and better patient
management.
Exercise Stress Testing
Patients are requested to fast for 4-6 hours prior to the
test to minimize splanchnic blood distribution and
exercise or pharmacologic stress-induced gastric dis-
tress. Cardiac medications may be held depending on
460 NUCLEAR MEDICINE: THE REQUISITES

Anterior

LAD

Septum LAD LCx Lateral

RCA

Inferior

Anterior

LAD
Figure 14-9 Two- and three-head gamma camera detector
configurations.A common configuration with the two-headed
LAD systems is with the heads at right angles to each other (lower).
Apex or Thus, maximal count rate is achieved over a 180-degree rotation.
RCA Three-heads allow for an even higher count rate and are typically
acquired over a 270- to 360-degree rotation.

RCA

Inferior

Apex

LAD or RCA

Lateral
Septum LCx
LAD

Figure 14-8 SPECT schematic correlation of myocardial wall Figure 14-10 Processing SPECT to obtain short and long axis
segments and vascular supply. Short-axis, vertical long-axis, and cross-sectional cuts. Schematic diagram displays the standard
horizontal long-axis SPECT views. LAD, Left anterior descending orientation along the short and long axis of the heart.
artery; LCx, left circumflex branch; RCA, right coronary artery.

the indication for the stress test, whether for diagnosis
or to determine the effectiveness of therapy ( Box 14-8).
Beta-blockers may prevent achievement of maximum
heart rate and nitrates or calcium channel blockers may
mask or prevent cardiac ischemia, limiting the test’s
ability to detect coronary disease. On the other hand,
assessing the adequacy of drug therapy in blocking
ischemia requires the patient to remain on the medica-
tion. A negative test while the patient is taking cardiac
medications augurs well from a prognostic standpoint.
The decision is left to the discretion of the referring
physician.
In addition to a standard 12-lead ECG baseline evalua-
tion and continuous monitoring during the test, an intra-
venous line with a keep-open solution is placed. Graded
treadmill exercise is performed ( Fig. 14-14) according
to a standardized protocol, such as Bruce protocol
( Table 14-3). When the patient has achieved maximal
exercise or peak patient tolerance,the radiopharmaceuti-
cal is injected. Exercise is continued for another 1-2 min-
utes to ensure adequate uptake reflecting the perfusion
pattern at peak stress. Early discontinuation of exercise
may result in tracer distribution reflecting perfusion at
submaximal exercise levels. Box 14-9 lists reasons for
 Cardiac System 461

Figure 14-11 Standard display of SPECT cross-sectional images correlated with cardiac anatomy.
The planes are cut along the short and long axis of the heart, top (short-axis), middle (vertical long
axis), bottom (horizontal long axis).The left ventricle is best seen due to its greater myocardial mass;
the right ventricle is less well seen and the display emphasizes the left ventricle.Atria are not
visualized.

terminating exercise. Many of these are manifestations
of ischemia and others are due to underlying cardiac or
pulmonary decompensation.
Some patients cannot perform exercise treadmill stress
because of medical problems such as pulmonary disease
or lower extremity musculoskeletal problems. Alternative
approaches to exercise stress have been used, including
isometric handgrip ( Box 14-10). However, pharmaco-
logic stress is the usual alternative to exercise.
Pharmacologic Stress Testing
Pharmacologic stress with dipyridamole ( Persantine),
adenosine ( Adenoscan), or dobutamine is used in up to
40% of patients referred for stress myocardial perfusion
studies.
Dipyridamole and Adenosine
Mechanism of Pharmacologic Effect Both adeno-
sine ( AdenoScan, Fujisawa) and dipyridamole (Persan-
tine, Bristol-Myers Squibb) are potent coronary vasodila-
tors capable of producing a threefold to fourfold increase
in normal coronary blood flow. Adenosine is normally
endogenously produced in coronary endothelial cells.
When released intravascularly, it activates coronary recep-
tors that produce vasodilation. Dipyridamole exerts its
pharmacologic effect by blocking the reuptake mecha-
nism of adenosine and raising endogenous adenosine
462 NUCLEAR MEDICINE: THE REQUISITES

Unlike exercise, this method is not a test of ischemia, but

Box 14-4 Scintigraphic Patterns for rather of coronary flow reserve. Regardless, compara-
Specific Vascular Distributions: tive studies have shown similar scintigraphic patterns
Stenosis and Obstruction and overall diagnostic accuracy for exercise and pharma-
cologic stress.
Coronary Artery Scintigraphic Perfusion Defects Patient Preparation Both adenosine and dipyri-
damole are antagonized by drugs and food containing
Left anterior Septum, anterior wall, apex chemically related methylxanthines, such as theophylline
descending and caffeine ( Fig. 14-15). They must be discontinued for
Left circumflex Lateral wall, posterior wall,
24 hours prior to the study since they may counteract
posterior inferior wall, apex
Right coronary Inferior wall, posterior inferior
the effectiveness of vasodilation ( Box 14-8). Adenosine
wall, right ventricular wall and dipyridamole can cause bronchospasm, thus these
Left main Anterior wall, septum, drugs should not be used in patients with a history of
coronary posterolateral wall severe bronchospastic disease.
Multiple-vessel Multiple vascular bed Methodology The technical details for dipyridamole
disease perfusion defects and adenosine infusion protocols differ ( Box 14-11 and
Post stress ventricular dilatation 14-12) because of their different pharmacokinetics
and increased Tl-201 lung uptake ( Table 14-5). Both are administered as slow infusions
of 4–6 minutes. A mild increase in heart rate and
mild reduction in blood pressure confirm the drug’s
blood levels. Adenosine and dipyridamole are used pharmacologic effect. Some protocols have the patient
interchangeably, although they have somewhat different perform mild exercise during infusion ( e.g., walking in
infusion protocols ( Boxes 14-11 and 14-12), incidence of place or handgrip isometric stress) to decrease the side
side effects ( Table 14-4), and methods to reverse side effects of hypotension and dizziness and to minimize
effects. liver activity seen on scintigraphy by diverting the blood
Because coronary vessels with significant stenoses flow to leg muscles.
cannot increase blood flow to the same degree as normal Side Effects Mild nausea, dizziness, headache, and
vessels, vasodilator stress results in vascular regions of flushing are common with adenosine and Persantine
relative hypoperfusion on myocardial perfusion scintigra- ( Table 14-4). Approximately 20–35% of patients experi-
phy, similar to that seen with exercise-induced ischemia. ence chest pain, although it is usually not ischemic in

Figure 14-12 Standardization of SPECT myocardial segments.This is the method recommended

and published in Circulation and the Journal of Nuclear Cardiology in an attempt to standardize
the regions of the myocardium into 17 regions for all cardiac imaging.The diagram also correlates
coronary artery anatomy with regional perfusion. However, some computer software systems use
a different number of regions. (Modified with permission from Cerqueira MD,Weissman J, Dilsizian V,
et al: J Nucl Card 2:240–245, 2002.)
 Cardiac System 463

origin. Rarely a coronary steal syndrome may produce

true ischemia. Dyspnea more commonly occurs with
adenosine and A-V conduction blocks are seen exclusively
with adenosine. If ECG ST-T depression occurs during
infusion, it is diagnostic for coronary disease.
With adenosine, no antidote is required to reverse
side effects. The infusion is simply terminated. Because
of adenosine’s short half-life in serum ( less than 10 sec-
onds), conduction blocks or other adverse symptoms
resolve promptly. However, because the action of
dipyridamole is often prolonged due to its 20–30
minute serum half-life, aminophylline is required to
Box 14-6 Stress Testing: Indications and
Contraindications
INDICATIONS
Diagnosis of coronary artery disease
Evaluation of known coronary disease; location and
extent of ischemia
Determine the cause for change in symptom pattern in
patients with known coronary artery disease
Evaluate the effectiveness of medical therapy
Risk stratification post-myocardial infarction
Preoperative evaluation for major noncardiac surgery in
patient with known coronary disease
Assessment after percutaneous transluminal coronary
angioplasty or coronary artery bypass grafting
Guide to rehabilitation therapy

CONTRAINDICATIONS
Acute myocardial infarction
Unstable angina
Severe tachyarrhythmias or bradyarrhythmias
Uncontrolled symptomatic heart failure
Critical aortic stenosis
Acute aortic dissection
Pulmonary embolism
Poorly controlled hypertension

Box 14-7 Reasons for Failing to Achieve

Adequate Exercise

Figure 14-13 Relationship between blood flow and severity of

Poor general conditioning, low exercise tolerance
coronary stenosis.At rest, myocardial blood flow is not reduced
until a coronary stenosis approaches 90%. It then begins to drop Poor motivation
off. However, with increased rates of coronary blood flow Arthritis, other musculoskeletal problems
produced by exercise or pharmacologic stress, less severe stenoses Lung disease
(50–75%) cause reduced coronary flow. Peripheral vascular disease
Medications (beta-blockers)
Angina
Arrhythmia
Box 14-5 Exercise Testing: Rationale and Cardiac insufficiency
Endpoint Measures
Box 14-8 Drugs That Interfere with
PHYSIOLOGIC RATIONALE
Stress Testing: Recommended
Physical exercise increases cardiac work Withdrawal Interval
Increased work increases myocardial oxygen demand
Normal coronary arteries dilate and flow increases
Stenotic vessels cannot dilate and flow reserve is limited Drug Withdrawal Interval
Myocardial ischemia is induced
EXERCISE
MANIFESTATIONS OF MYOCARDIAL ISCHEMIA Beta-blockers 72 hr
Electrocardiogram: Ion flux across cell membrane is Calcium channel blockers 48–72 hr
impaired, produces ST segment depression on ECG Nitrates (long acting) 12 hr
Myocardial perfusion imaging: Decrease in regional flow
PHARMACOLOGIC
produces cold defect area on scintigraphy
Radionuclide ventriculography: Regional wall motion Aminophylline 36 hrs
abnormality and/or fall in LVEF Caffeine 24 hrs
464 NUCLEAR MEDICINE: THE REQUISITES

Box 14-9 Indications for Terminating

a Stress Test

Patient’s request
Inability to continue due to fatigue, dyspnea, or faintness
Moderate to severe chest pain
Dizziness, near syncope
Pallor, diaphoresis
Ataxia
Claudication
Ventricular tachycardia
Atrial tachycardia or fibrillation
Onset of second- or third-degree heart block
ST segment depression >3 mm
Decrease in systolic blood pressure from baseline
Increase in systolic blood pressure above 240 mm Hg or
diastolic above 120 mm Hg

Box 14-10 Cardiac Stress Tests: Exercise

and Nonexercise

LEG EXERCISE
Treadmill
Figure 14-14 Treadmill exercise.Treadmill graded exercise Bicycle ergometer
with ECG, blood pressure, and heart rate monitoring.
OTHER CARDIAC STRESS TESTS
Isometric handgrip exercise
Cold pressor test
Esophageal pacing

PHARMACOLOGIC STRESS
Dipyridamole, adenosine, dobutamine
Table 14-3 Treadmill Exercise Stress Test

Duration Total Speed Box 14-11 Dipyridamole Pharmacologic

Stage (min) time (mile/h) Grade (%) Stress: Protocol
STANDARD BRUCE PROTOCOL
Time from
1 3 3 1.7 10
Start of
2 3 6 2.5 12
3 3 9 3.4 14
Infusion (min) Protocol
4 3 12 4.2 16
Obtain baseline ECG and blood
5 3 15 5.0 18
6 3 18 6.0 20 pressure; report at 1-min intervals
0–4 Administer dipyridamole 0.56 mg/kg
MODIFIED BRUCE PROTOCOL for 4 min intravenously
1 3 3 1.7 0 6–8 Inject radiopharmaceutical intra-
2 3 6 1.7 5 venously 3–4 min after
3 3 9 1.7 10 dipyridamole infusion
4 3 12 2.5 12 20 Begin imaging at 14 minutes after
5 3 15 3.4 14 dipyridamole infusion completed
6 3 18 4.2 16
Optional as a Administer 75 to 100 mg
7 3 21 5.0 18
routine or as aminophylline slowly by
needed for intravenous injection
The modified Bruce starts with the same speed as the standard Bruce, but with
no slope, followed by slight increase in slope, and then in speed.This protocol is adverse
suited for elderly patients or patients where one anticipates difficulties with symptoms
physical performance.
 Cardiac System 465

Box 14-12 Adenosine Pharmacologic

Stress: Protocol

Time from
Start of
Infusion (min) Protocol

Obtain baseline ECG and blood

pressure; report at 1-min intervals
0–6 Administer adenosine
0.14 mg/kg/min for 4–6 min
intravenously
3 Inject radiopharmaceutical
intravenously at 3 min from
start of infusion
9–10 Begin imaging 3 minutes after
adenosine infusion completed

Figure 14-15 Close chemical relationship of adenosine and

dipyridamole to theophylline and caffeine.

Table 14-4 Adverse Effects: Adenosine and

Dipyridamole

Adverse effect Adenosine Dipyridamole Table 14-5 Comparison of Pharmacologic

Myocardial Stress Agents
Flushing 37% 3%
Dyspnea 35 3
Chest pain 25 20 Adenosine Dipyridamole Dobutamine
Gastrointestinal symptoms 15 6
Headache 14 12 Half-life <10 sec 30–60 min 2 min
Dizziness 9 12 Mean time to 55 sec 6.5 min 10 min
A-V block 8 2 peak coronary
ST-T wave changes 6 8 flow velocity
Arrhythmia 3 5 Onset of action seconds 2 min 1–2 min
Hypotension 2 5 Side effects 0.6% 16% 5%
Bronchospasm 0.1 0.15 requiring
Myocardial infarction 0.0001 0.05 medical
Death 0 0.5 intervention

reverse its side effects. Some clinics routinely adminis-
ter aminophylline at the end of dipyridamole stress to
prevent and terminate its common side effects.
Accuracy Comparative studies have reported similar
accuracy for detection of significant coronary disease for
adenosine and dipyridamole compared to exercise-stress
SPECT myocardial perfusion scintigraphy. The
disadvantage is the lack of information on functional
cardiac status provided by exercise.
Dobutamine
For patients unable to exercise but with contraindica-
tions to dipyridamole and adenosine (e.g., asthma), dobu-
tamine can be used as an alternative.
Mechanism of Action This synthetic catecholamine
acts on alpha- and beta-adrenergic receptors producing
inotropic and chronotropic effects that increase cardiac
work. In normal coronary arteries,this results in increased
blood flow. In the face of significant stenosis,regional flow
does not increase,resulting in scintigraphic patterns similar
to that seen with exercise and pharmacologic stress.
Methodology The infusion begins with 5 μg/kg/min
infused for 3 minutes, then increased to 10 μg/kg/min for
another 3 minutes and increased by that amount every
10 minutes until a maximum of 40 μg/kg/min is achieved.
The radiopharmaceutical is injected one minute after the
maximal tolerable dose and the dobutamine infusion
continued for at least 1 minute.
Accuracy Dobutamine perfusion imaging is reported
to have accuracy similar to exercise or dipyridamole/
adenosine stress. The major limitation is the frequent
466 NUCLEAR MEDICINE: THE REQUISITES

Box 14-13 Protocols for Stress Myocardial Box 14-14 Advantages and Disadvantages
Perfusion Scintigraphy of Tc-99m Sestamibi/
Tetrofosmin for Myocardial
Perfusion Imaging
Radiopharma-
Method ceutical Rationale
ADVANTAGES
2-day Tc-sestamibi/ Obesity,
Higher count rates; better quality SPECT images and
tetrofosmin image
gated SPECT possible
quality
Higher energy photons; fewer attenuation artifacts
1-day Tc-sestambi/ Image
Simultaneous assessment of perfusion and function
tetrofosmin quality,
First-pass assessment of right and left ventricular
efficiency
function possible
1-day Tl-201 Time tested,
viability DISADVANTAGES
Dual-isotope Tl-201 and Image
No redistribution
Tc-sestamibi/ quality,
Lung uptake not diagnostic
tetrofosmin viability,
Less extraction at hyperemic flows
logistics
Less sensitive than Tl-201 for viability assessment

occurrence of side effects (e.g., chest pain and

arrhythmias) and the inability of many patients to tolerate
the maximum required dose.
One- and Two-Day Protocols
Various different stress perfusion imaging protocols are
used, depending on the radiopharmaceutical used,
patient size, logistics of the clinic, and preference of the
physicians ( Box 14-13).
Thallium-201
Originally, a two-day protocol was used for Tl-201, with
separate rest and stress injections. It was soon appreci-
ated that Tl-201 redistributes, allowing for assessment of
both stress and rest perfusion following a single injection
on the same day. After stress,initial images are obtained at
10–15 minutes and delayed images at 3 hours ( Box 14-2,
Fig. 14-B [see color insert]). Although the new Tc-99m
radiopharmaceuticals are increasingly used for myocardial
perfusion scintigraphy,Tl-201 is still preferred by some
because of its lesser liver and bowel activity that can
adversely impact interpretation. It is also used for the diag-
nosis of myocardial viability,discussed later.
Tc-99m Sestamibi and Tc-99m Tetrofosmin
Because these radiopharmaceuticals do not redistribute
or washout significantly from the myocardium over time,
two separate injections are required. A summary of the
advantages and disadvantages of Tc-99m sestamibi and
tetrofosmin compared to Tl-201 are listed in Box 14-14.
Several different approaches have been used.
One-Day Both rest and stress studies can be
performed on the same day. The patient receives a lower
administered dose for the first study (8–10 mCi [266–370
MBq]) and a several-fold higher dose ( 25–30 mCi
[925–1110 MBq]) for the second study. The most
common approach is to do the rest study first followed by the
stress study. The lower dose study is more susceptible to
attenuation effects similar to that seen with Tl-201. The
second study commences approximately 1.5 hours later to
allow time for decreased background activity biological
clearance and decay (Fig.14-C [see color insert]).
Two-Day Tissue attenuation can be marked in large
patients and can result in image poor quality, particularly
with the lower dose study. A 2-day approach minimizes
this problem allowing administration of the maximum
dose (25–30 mCi) for both studies. This approach is most
commonly used in obese patients.
Dual Isotope This approach takes advantage of the
different photopeaks of the Tc-99m agent ( 140 keV) and
Tl-201 (69–83 keV) (Fig.14-A1 [see color insert]). Simultane-
ous acquisition (rest Tl-201 and stress Tc-99m sestamibi or
tetrofosmin) is desirable for efficiency reasons;however,it is
not practical because downscatter of Tc-99m into the Tl-201
window is significant. Thus,theTl-201 rest study is performed
first using 3.0–3.5 mCi (110–130 MBq),followed by the stress
study,using 20–30 mCi (740–1110 MBq) of the Tc-99m agent.
Upscatter of the higher energy Tl-201 photons (167 keV) is
minimal because of their low abundance (10%).
The dual isotope protocol has two advantages. The
study can be completed more rapidly than Tc-99m pro-
tocols because imaging begins earlier, soon after injec-
tion for the rest study and no delay is required before
starting the stress study. Another advantage is that Tl-201
can provide information on viability ( hibernating
myocardium).
Image Interpretation
Coronary Anatomy and Myocardial Perfusion
Although the anatomy of the coronary circulation
varies in its details, the distribution of the major vessels

is reasonably predictable ( Fig. 14-16, Box 14-4). The left
anterior descending coronary artery serves most of the
septum and the anterior wall of the left ventricle. Its
diagonal branches course over the lateral wall and sep-
tal perforators penetrate into the septum. The left cir-
cumflex coronary artery and its marginal branches
serve the lateral and posterior walls. The right coro-
nary artery and its posterior descending branch serves
the right ventricle, the inferior portion of the septum,
and portions of the inferior wall of the left ventricle.
Figure 14-16 Normal distribution of the coronary arteries.The
left main coronary artery is only 0–15 mm in length before
dividing into the left anterior descending and left circumflex
arteries. Important branches of the left anterior descending artery
are the diagonal and septal branches.The left circumflex artery
(LCFx) has obtuse marginal branches.The right coronary artery
(RCA) originates separately and has important branches that
include the posterior right ventricular branches, the posterior
descending artery (PDA).“Dominance”refers to which coronary
artery (RCA or LCFx) supplies the diaphragmatic surface of the left
ventricle and posterior septum by giving rise to the posterior
descending and posterior left ventricular branches. In 85% of
patients the RCA is dominant, in 7–8% the LCFx is dominant and in
7–8% there is balanced circulation between the two. Right
posterior descending (RPD), right marginal branch artery (RMA).
Cardiac System 467
The apex may be perfused by branches from any of the
three main vessels.
Quality Control of SPECT Acquisition
Patient Motion Movement of the patient during the
study can degrade image quality. A review of raw
acquisition data should be routine. This is best done by
displaying all image projections in an endless loop
cinematic rotating display. Greater than 2-pixel deviation
can adversely affect image quality.
Another method for confirming motion is to review
the “sinogram” display. Each projection image is stacked
vertically. Each is compressed, with maintenance of the
count density distribution in the X-axis but minimization
of the count density distribution in the Y-axis. Because
the heart is not in the center of the camera radius of rota-
tion, the position of the left ventricle in the stacked
frames varies sinusoidally. Any significant motion will be
seen as a break in the sinogram (Fig. 14-17).
If there is significant motion, the study should ideally
be reacquired. If this is not possible, motion correction
programs are available with most camera computer sys-
tems. However, these programs correct only in the verti-
cal axis.
Image Quality Camera quality control is critical
and discussed in Chapters 3 and 4. Image quality may be
poor due to insufficient activity in the heart,due either to
low dose ( e.g., if much of the dose was inadvertently
injected and retained subcutaneously) or the result of
Figure 14-17 Sinogram to detect for motion. Projection images
are stacked vertically. Because the heart is not in the center of the
camera radius of rotation, the position of the left ventricle in the
stacked frames varies sinusoidally.Any significant motion will be
seen as a break in the sinogram. Note the horizontal break in the
mid-sinogram (left).The motion corrected (right).
468 NUCLEAR MEDICINE: THE REQUISITES
soft tissue attenuation (discussed later). The presence of
free Tc-99m pertechnetate will increase background and
degrade image quality.
Normal Myocardial Perfusion Scintigraphy
Rest SPECT Images Normal myocardial perfusion
rest images with Tl-201 and Tc-99m sestamibi/tetro-
fosmin show similar uniform uptake throughout the left
ventricular myocardium (Figs.14-5 and 14-6),although the
Tc-99m agents usually have superior image quality. The
right ventricle is seen to a lesser extent because of its
smaller myocardial muscle mass. Right ventricular
hypertrophy results in increased uptake (Fig.14-18). Atria
are not visualized.
On short-axis SPECT views, the left ventricle has
a doughnut appearance. The lateral wall usually appears to
have more uptake than the anterior or inferior wall.
Decreased uptake is seen near the base of the heart in the
region of the membranous septum (Fig. 14-6). The valve
planes have an absence of uptake, giving the heart a horse-
shoe or U-shaped appearance on long-axis ( coronal and
Figure 14-18 Right ventricular hypertrophy. Patient with interstitial fibrosis and severe
pulmonary hypertension. Stress and rest Tc-99 sestamibi show prominent uptake in a hypertrophied
right ventricle.
sagittal) SPECT slices. The normal myocardium appears
thinner at the apex (“apical thinning”). Mild lung uptake is
seen. Increased resting lung uptake occurs in heavy smok-
ers, patients with underlying lung disease, and those with
congestive heart failure (Fig.14-19).
Stress SPECT Images Normal myocardial perfusion
images obtained after exercise or pharmacological stress
are not strikingly different in distribution from those
obtained at rest; however, there are differences. The
cardiac target-to-background ratio of the stress images
is higher due to increased cardiac uptake. Right
ventricular uptake is often relatively increased, but still
considerably less than the left ventricle. With treadmill
exercise, less activity is seen in the liver because of
diversion of blood flow from the splanchnic bed to leg
muscles. Assessing the degree of uptake in the liver can
serve as an internal quality control check on the
adequacy of exercise. Pharmacological stress with
adenosine and dipyridamole stress results in considerable
liver activity.
 Cardiac System 469

Figure 14-19 “Rest-rest” Tl-201 scintigrams: hibernating myocardium.This patient had prior
myocardial infarctions and heart failure. (Top row) Initial rest images show decreased uptake in the
septum, apex, and anterior wall.The lung shows marked uptake, compatible with congestive heart
failure. (Bottom row) On the 4 hour delayed rest images, tracer has accumulated to a variable degree
in all areas of initial abnormality.The inferoapical defect on the left anterior oblique view has filled in
almost completely (arrows). Consistent with chronic ischemia (hibernating myocardium), not
myocardial infarction.

Planar Imaging The appearance of the heart
depends to some extent on the patient’s habitus and the
orientation of the heart in the chest (e.g., vertical or
horizontal). The heart has a variably circular or ellipsoid
shape in different views ( Figs. 14-4 and 14-5). The right
ventricle is not usually seen on planar studies with Tl-201
but is often seen with Tc-99m agents. When present,
lung uptake is easily appreciated on planar imaging
( Fig. 14-19) or the raw projection ( planar) images of
SPECT ( Fig. 14-20).
Attenuation Artifacts Different patterns of soft
tissue attenuation are seen in males and females.
Males typically have decreased activity in the inferior
wall ( Figs. 14-4, 14-C, 14-D [see color insert]). This is
due to diaphragmatic attenuation, meaning attenuation
by subdiaphragmatic organs interpositioned between
the heart and gamma camera. The amount of attenua-
tion effect is dependent on patient’s size, shape, and
anatomy.
Females often have relatively decreased activity in the
anterior wall, apex, or lateral portion of the heart, second-
ary to breast attenuation, dependent on the size and posi-
tion of the breasts ( Fig. 14-21B). Women also have subdi-
aphragmatic attenuation, but breast attenuation is
dominant. Anterior attenuation can be seen occasionally
in large males.
Other causes of soft tissue attenuation include excessive
adipose tissue or muscle hypertrophy and inability to raise
the left arm during imaging because of arthritis,fracture,or
lymphedema. Attenuation artifacts can be anticipated by
routinely reviewing the cinematic rotating raw data display
(Fig. 14-21A). Attenuation artifacts are generally worse for
Tl-201 studies because of the lower photon energy, but are
routinely seen with Tc-99m agents (Fig.14-4).
Myocardial perfusion artifacts due to breast or subdi-
aphragmatic attenuation seen at both stress and
rest could potentially be misinterpreted as myocardial
infarction. Alternatively, if the breasts are in somewhat
different positions for the two studies, this might be
interpreted as ischemia. Effort should be made to ensure
similar breast position for both sets of images ( e.g., both
studies with or without bra or breast binder).
470 NUCLEAR MEDICINE: THE REQUISITES
Figure 14-20 Stress-induced Tl-201 lung uptake.The lung uptake is caused by stress-induced
cardiac decompensation and is usually associated with three vessel coronary artery disease.This
patient also had transient ischemic dilation (see Fig. 14-E1 [see color insert]).
Attenuation Correction Many SPECT computer
systems now have attenuation correction capability for
cardiac studies, although some controversy exists
regarding its utility. A transmission map is acquired,
often with a gadolinium-153 gamma source rotating
around the patient. Some systems allow simultaneous
acquisition of emission and transmission data. Hybrid
SPECT-CT systems use an x-ray tube source for the
transmission map to correct for attenuation ( Figs. 14-C
and 14-D [see color inserts]). When attenuation correct
images are obtained, nonattenuation-corrected images
should still always be reviewed. Misregistration of the
emission and transmission images is a potential problem
with the hybrid SPECT-CT systems. Scatter correction
methods are on some camera computer systems;
however, most require further validation.
Gated SPECT is routine in many nuclear medicine clin-
ics ( Figs. 14-22 to 14-24). The gated SPECT is typically
performed poststress. It can be helpful in differentiating
attenuation artifact from infarction. With good wall
motion and myocardial thickening, the decreased activity
is likely due to attenuation. Prone imaging has also been
used to differentiate attenuation affect from infarction.
Extracardiac Uptake
Myocardial perfusion radiopharmaceuticals are taken up
in all cellular, metabolically active tissues in the body
except for the brain, in which they do not cross the nor-
mal blood brain barrier. Structures accumulating Tl-201
or Tc-99m sestamibi/tetrofosmin that may be in the field
of view are the thyroid, salivary glands, skeletal muscle,
and kidneys. Prominent liver uptake is usual in rest and
with pharmacologic stress studies. Gallbladder and
intestinal activity are commonly seen.
Activity from the liver and bowel may cause scatter
into the inferior and adjacent walls of the heart, increas-
ing the apparent uptake, which can complicate inter-
pretation. Focal hot subdiaphragmatic activity adjacent
to the heart can also produce apparent cold defects due
to filtered backprojection reconstruction artifacts.
All three radiopharmaceuticals are taken up in benign
and malignant tumors, thus incidental uptake in a known
or unknown tumor is not rare ( Fig. 14-25) and should be
noted and reported when present.
Diagnostic Patterns in Coronary Artery Disease
Diagnosis of Ischemia and Infarction Terms
used to characterize the status of myocardium ( e.g.,
ischemia, infarction, hibernating, stunned) are defined in
Box 14-15. The diagnostic schema presented in Table 14-6
uses the appearance of the scans on the stress and rest
( or redistribution) studies to characterize myocardial
perfusion as:
1. Normal: no defects noted on either image set ( Figs.
14-5 and 14-6)
2. Ischemia: cold defects on poststress images that fill
in on delayed images ( Figs. 14-A, 14-B, 14-E, 14-F [see
color inserts], 14-26)
3. Infarction: defects that remain “fixed” between the
image sets ( Figs. 14-27 and 14-G [see color insert])
 Cardiac System 471

Figure 14-21 Breast attenuation. A, Single raw data acquisition projection of a cinematic display
at stress (left) and rest (right) illustrates the artifact produced by breast attenuation. Note the
apparent decreased activity in the upper portion of the heart. B, The SPECT cross-sectional slices of
the same patient show moderate anterior wall attenuation best seen on the short axis and vertical
long axis views.
472 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-22 Wall motion, thickening, and bulls-eye analysis. End-diastolic and end-systolic
perfusion from different reconstructed projection angles.This bull’s eye display portrays perfusion
at end-systole and end-systole as well as wall thickening on gated SPECT.The top row of images
represents end-systole and the bottom row end-diastole.An area of relatively diminished tracer
uptake inferolaterally corresponds with decreased wall thickening (dark area) on the bull’s eye
display.

Figure 14-23 Gated SPECT perfusion scan in a normal subject.Left ventricular ejection fraction is 65%.
 Cardiac System 473

Figure 14-24 Gated SPECT in a patient with coronary artery disease. Decreased tracer uptake is
seen in multiple wall segments.The left ventricular cavity is dilated with an estimated end-diastolic
volume of 168 ml.The LVEF is 23%.

Figure 14-25 Incidental tumor uptake on stress and rest Tc-99m sestamibi studies.
A, Bilateral breast masses are seen on the cinematic display of raw data. Diagnosis of breast cancer.
B, Thymoma.
474 NUCLEAR MEDICINE: THE REQUISITES

Box 14-15 Definitions Describing the Status of the Myocardium

Term Definition and Scan Appearance

Myocardial ischemia Oxygen supply below metabolic requirements due to inadequate blood circulation as a result
of coronary stenosis
Photon deficient (cold defect) on stress perfusion scintigrams
Myocardial infarction Necrosis of myocardial tissue, as a result of coronary occlusion
Photon deficient on rest and stress perfusion and metabolic imaging studies
Transmural infarction Necrosis involves all layers from endocardium to epicardium
High sensitivity for detection by perfusion imaging
Subendocardial infarction Necrosis involves only muscle adjacent to endocardium
Lower sensitivity for detection on perfusion imaging
Myocardial scar Late result of infarction; photon deficient on scintigraphy
Hibernating myocardium Chronically ischemic myocardium with decreased blood flow and down regulation
of contractility; reversible with restoration of blood flow
No perfusion on rest imaging, poor ventricular contraction
Improved perfusion on rest/rest or reinjection Tl-201 imaging
Increased uptake by FDG metabolic imaging compared to perfusion scan
Stunned myocardium Myocardium with persistent contractile dysfunction despite restoration of perfusion after
a period of ischemia; usually improves with time
Normal by perfusion imaging, poor ventricular contraction
Uptake by FDG metabolic imaging

Patients may have a combination of fixed and tran-
sient defects ( Figs. 14-28 and 14-G [see color insert]).
After initial assessment of the presence or absence of
perfusion defects, a complete evaluation of the stress
study includes assessment of the location, size, severity,
and likely vascular distribution of the visualized abnor-
malities. Various computer quantitative methods can be
used in conjunction with image analysis ( Figs. 14-29,
14-30, 14-A2, 14-B2, 14-E2, 14-F2, and 14-F3 [see color
inserts]).
Perfusion defects caused by coronary artery disease
are more common distally, rather than at the base of the
heart. A true perfusion defect should be seen on more
than one cross-sectional slice. Certainty also increases
with lesion size and the degree or severity of photon
deficiency.
Left Bundle Branch Block (LBBB) Stress-induced
reversible hypoperfusion of the septum can be seen in
patients with LBBB in the absence of angiographic
coronary disease. The apex and anterior wall are not
involved, as would be expected with left anterior
descending coronary artery disease. The stress-induced
decreased septal blood flow is thought due to asynchronous
relaxation of the septum, which is out of phase with
diastolic filling of the remainder of the ventricle when
coronary perfusion is maximal. This scintigraphic
abnormality is not seen with pharmacologic stress, thus
dipyridamole or adenosine stress is indicated for patients
with LBBB and probably also for patients with ventricular
pacemakers.
Poor Prognostic Findings on Perfusion Scintigraphy
Multiple Perfusion Defects The presence of perfu-
sion defects in more than one coronary artery distri-
bution area points to multiple vessel disease. The more
perfusion defects and the larger their size, the worse is
the prognosis.
Not all significant coronary artery stenoses are seen
on perfusion scintigraphy. Stress-induced ischemia of
the most severe stenotic lesion limits further exercise
and thus other stenoses may not be seen scintigraphi-
cally. Because interpretation is based on relative perfu-
sion of adjacent walls, multiple vessel disease may be
underestimated. Three-vessel balanced disease may not
be seen at all. Two findings described later are highly
associated with multivessel disease.
Transient Ischemic Dilatation Although the nor-
mal response of the heart is to dilate during stress,
the SPECT acquisition is poststress. The normal heart
will return to normal size promptly after exercise.
Persistent dilatation poststress is abnormal and indi-
cates multivessel disease ( Fig. 14-E [see color insert]).
The persistent dilatation is due to myocardial stunning
during stress.
Tl-201 Lung Uptake Tl-201 lung uptake induced by
exercise is also a poor prognostic sign caused by stress-
induced heart failure manifested by ventricular dysfunction,
 Cardiac System 475

elevated left ventricular end-diastolic pressure and pul- are usually due to myocardial infarction. However,
monary capillary wedge pressure ( Fig. 14-20). Lung-to- approximately 20% of patients with these findings do not
myocardial activity ratios greater than 0.5 are abnormal. have infarction, but rather severe chronic ischemia or
Lung uptake with Tc-99m sestamibi/tetrofosmin is less hibernating myocardium ( Box 14-15, Fig. 14-31A).
reliable because of normal lung activity. Although severely underperfused, the myocytes have
Myocardial Viability (Hibernating Myocardium) preserved cell membrane integrity and sufficient meta-
The scintigraphic findings of regional hypoperfusion at bolic activity to maintain cellular viability, but not con-
stress and rest associated with myocardial dysfunction tractility. Those patients with viable but hypoperfused
myocardium may benefit from coronary revasculariza-
tion with improvement in cardiac function and reduc-
tion in annual mortality from 16% to 3%.
Echocardiography, gated blood pool ventriculography,
Table 14-6 Diagnostic Patterns: Stress Myocardial or gated SPECT cannot make the distinction between
Perfusion severe ischemia and scar because the severely ischemic
segments demonstrate reduced or absent contractility.
Immediate Resting delayed These segments are “hibernating” in a functional and
poststress or reinjection Diagnosis metabolic sense.
Tl-201 uptake is an energy-dependent process requir-
Normal Normal Normal
Defect Normal Ischemia ing intact cell membrane integrity, thus uptake implies
Defect Defect (unchanged) Infarction* preserved myocardial cell viability. The magnitude of up-
Defect Some normalization Ischemia and scar* take correlates with the extent of tissue viability. The use of
with areas of F-18 FDG to diagnose viability, considered the gold stan-
persistent defect
dard,is discussed later in the cardiac PET section. However,
Normal Defect “Reverse”redistribution
various protocols utilizing Tl-201 have been successfully
*Delayed Tl-201 imaging without reinjection may overestimate the presence and used to differentiate viable or hibernating myocardium
amount of infarcted area because of incomplete redistribution. from myocardial infarction and are reviewed here. Tc-99m

Figure 14-26 Exercise induced ischemia. SPECT stress-rest images with Tc-99m sestamibi.
Decreased tracer uptake in the anterior wall is best seen on the long-axis views (middle rows).The
defect substantially reperfuses on the resting images.This pattern indicates exercise-induced
ischemia and coronary artery disease.
476 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-27 Large inferior wall infarction. SPECT images obtained at stress and rest in a patient
with a history of prior myocardial infarction.A large fixed defect involving the inferior wall of the
heart can best be seen on the short-axis and vertical long-axis views. No substantial change in the
scintigraphic appearance is noted between the poststress and rest images.

Figure 14-28 Combined ischemia and infarction. SPECT images obtained at stress and rest with
Tc-99m sestamibi reveal a large perfusion defect in the anterior wall best seen on the vertical long-
axis views (arrows) that reperfuses on rest images.A fixed perfusion defect is present in the inferior
wall and is best seen on the short-axis views.This pattern of both fixed and transient defects is
indicative of combined myocardial ischemia and scarring in a patient with multivessel disease.
 Cardiac System 477

Horizontal Long Axis Slices

Vertical Long Axis Slices

Short Axis Slices

A
Figure 14-29 Three-dimensional quantitative display of myocardial infarction. A, SPECT images
of a patient with a large fixed defect involving the apex. (The apex is at the bottom of the image on
the horizontal long-axis slices and to the left on the vertical long-axis slices.) The fixed defect is seen
on all three slice orientations.
Continued

sestamibi and tetrofosmin underestimate myocardial via-
bility and are not generally useful for this purpose.
Tl-201 Protocols for Assessing Myocardial Via-
bility Tl-201 redistribution occurring by 3-4 hours after
injection is dependent on several factors that include the
presence of viable myocytes, the severity of the initial
defect poststress, and the concentration and rate of
decline of Tl-201 in the blood ( see Fig. 14-1). Various
protocols have been used to detect viable myocardium
based on an understanding of these pharmacokinetics.
Late Redistribution Imaging One approach is to
allow additional time for redistribution to occur (e.g.,
8 to even 28 hours after the stress Tl-201 injection).
Severely ischemic myocardium, with slow uptake and
clearance, requires a longer time to redistribute.
Improvement in uptake has good positive predictive
value for identifying regions with potential functional
improvement. However, the negative predictive value is
suboptimal because of poor image quality caused by
a low count rate from radiotracer decay and biological
clearance from the body. Acquisition time can be
increased to improve the count rate; however, this
increases the likelihood of patient motion.
Tl-201 Reinjection This approach assumes that the
lack of redistribution is the result of a low Tl-201 blood
level. To increase the blood level, Tl-201 is reinjected
(usually 50% of the initial dose) after the redistribution
images are complete and then repeat images are
obtained 15–20 minutes later ( Fig. 14-32, Fig. 14-E [see
color insert]). Alternatively, some protocols routinely
478 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-29, cont'd B, Three-dimensional quantitative analysis of the images from A reveals
the large apical and inferior defect when the stress scintigrams are compared with a normal data set
(left images).The left set of images compares the patient’s data to a reference data set, and the right
set compares post-stress with rest testing.The fixed nature of the scintigraphic defect is illustrated
on the right. No differences are detected between rest and stress views.

reinject Tl-201 prior to the usual redistribution images.
Definite uptake is predictive of improvement in regional
left ventricular function after revascularization. The
presence of a severe persistent Tl-201 defect after reinjec-
tion identifies areas with a very low likelihood for
improvement in function.
One rationale for using Tl-201 for the rest study in
dual-isotope studies is the opportunity to get next day
imaging, with or without reinjection, in order to deter-
mine viability of a fixed defect.
Rest-Rest Tl-201 Redistribution When viability, not
inducible ischemia, is the clinical question, a rest-rest
study may be all that is needed ( see Fig. 14-19). The his-
tory of such a patient is that of known coronary artery
disease, prior myocardial infarction, and ventricular dys-
function, being considered for bypass surgery. In the
region of infarction, considerable viable chronically
ischemic myocardium may exist. The clinical question is
whether there is enough viable myocardium to justify
revascularization. After tracer injection at rest, images
are obtained 15 minutes later, which reflect regional
blood flow. Significant uptake on repeat rest images
3–4 hours later reflects viability and likely benefit from
revascularization.
Stunned Myocardium
After a transient period of severe ischemia followed by
reperfusion, there may be a state of delayed recovery of
regional left ventricular function. This is called stunned
myocardium ( Fig. 14-31B). The ischemic episode may
be single, multiple, brief, or prolonged, but not severe
 Cardiac System 479

Horizontal Long Axis Slices

Vertical Long Axis Slices

Short Axis Slices

A
Figure 14-30 Three-dimensional quantitative display: ischemia. A, SPECT images at stress and
rest in a patient with a large reversible defect involving the anterior wall with extension to the apex.
The defect is best seen on the vertical long-axis (middle row) and short-axis views.The rest images
are normal.
Continued

enough to cause necrosis. Tissue in the affected perfu-
sion watershed is viable and accumulates radiopharma-
ceutical immediately after reperfusion. The uptake of
tracer indicates viability, but the myocardial segment may
be akinetic or stunned. If the segment is only stunned
and not infarcted, wall motion will improve with time.
Stunning may be seen after thrombolysis or angioplasty
in patients who have had acute coronary occlusion.
Stress-induced transient ischemic dilation, stress-induced
Tl-201 lung uptake, and poststress gated SPECT Tc-99m
sestamibi/tetrofosmin ventricular dysfunction are all mani-
festations of stunned myocardium.
Reverse Redistribution
Worsening of a perfusion defect or the development of
a new defect on Tl-201 redistribution images compared
with immediate poststress images is called reverse redis-
tribution. It may also be seen with Tc-99m agents. The
reported causes are numerous and the clinical and diag-
nostic importance uncertain.
Reverse redistribution has been reported in severe
coronary artery disease, perhaps caused by differential
washout between normal and diseased areas. It is also
seen in some patients after infarction and after successful
thrombolytic therapy producing patency of the infarct-
related artery. If the latter situation, it is thought to be
due to imbalance in tracer delivery (perfusion) versus
the ability of stunned myocardium to retain the tracer,
leading to a high differential washout rate from the infarct
zone compared with periinfarct myocardial tissue.
However, reverse redistribution is neither sensitive nor
specific for coronary disease. There are reports of
reverse redistribution in a variety of seemingly unrelated
480 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-30, cont'd B, Three-dimensional quantitative analysis of the images in A reveals the
large stress-induced defect when the patient’s images are compared with a normal reference data set
(left images).The transient nature of the defect is illustrated in the images on the right where the
rest and stress images from the patient are compared.The three-dimensional analysis demonstrates
the extent of abnormality comparing the defect versus the normal reference data set (left) and
nicely shows the reversibility (right).

clinical situations such as after bypass surgery,postcardiac
transplantation, with Wolff-Parkinson-White syndrome,
sarcoidosis, Kawasaki disease, and Chagas’ disease.
Quantitative Analysis
A number of techniques are available for quantitative
analysis of myocardial perfusion scans. These tech-
niques typically use a normal database that provides a ref-
erence for the expected range of relative regional uptake
and/or washout rates. A commonly used method is the
polar map. Relative perfusion is presented in a two-
dimensional “bull’s-eye” display that is generated by map-
ping of circumferential profiles obtained from the
short-axis SPECT views, with the apex at the center of
the display and the base of the ventricle at the periphery
( Figs. 14-A2, 14-C, 14-F2, 14-G2, [see color inserts] and
14-19).
New approaches to quantitative analysis include three-
dimensional quantitative displays of regional perfusion,
difference displays in perfusion between rest and stress,as
well as quantitative analysis of wall motion and wall thick-
ening ( Figs. 14-A2, 14-C, 14-F2, 14-G2, [see color inserts],
and 14-19). Wall thickening derives from regions of inter-
est placed systematically around the myocardium ( Figs.
14-22 to 14-24 and 14-H [see color insert]) and the left ven-
tricular ejection fraction (LVEF) is obtained by measuring

Figure 14-31 A, Hibernating myocardium (chronic ischemia)
develops over time due to chronic hypoperfusion and causes
regional wall motion dysfunction (A).When perfusion is
reestablished by surgical intervention, myocardial function
returns, but gradually (B). B, Stunned myocardium is often due to
an acute ischemic episode such as thrombosis (A), which when
relieved with, for example, angioplasty (B), results in prompt
reperfusion. Functional recovery is considerably delayed (C).
Modified with permission from Dilsizian V and Narula J. Atlas of
Nuclear Cardiology, Current Medicine, Philadelphia, 2003.
the change in size of the ventricular cavity through the car-
diac cycle ( Figs. 14-23A and 14-24) using edge detection
algorithms.
The quantitative approach has some potential pitfalls,
including problems of misregistration with the reference
data set, use of data sets generated from other laborato-
ries on equipment different from that used in the
patient’s examination or on different patient popula-
tions, and lack of uniformity in the amount of exercise or
stress achieved. Nonetheless, the use of quantitative
methods has become standard practice.
Sensitivity and Specificity
Coronary angiography has been the gold standard for
determination of myocardial perfusion scintigraphy
accuracy. However, the estimated angiographic percent
coronary stenoses often do not correlate well with their
functional severity determined by coronary flow
reserve. The amount and degree of coronary disease is
often underestimated by angiography. Furthermore,
a true physiologic decrease in blood flow may sometimes
Cardiac System 481
Figure 14-32 Pharmacokinetics of Tl-201 reinjection to
diagnose hibernating myocardium. During routine Tl-201 imaging,
delayed images may show a persistent fixed perfusion defect in a
region of severe or chronic ischemia. Low blood levels of Tl-201 do
not allow for adequate redistribution of Tl-201 to the myocardium.
Augmenting the Tl-201 blood levels permits greater myocardial
uptake and scintigraphic evidence for redistribution and thus
viable myocardium. Modified with permission from Dilsizian V and
Narula J. Atlas of Nuclear Cardiology, Current Medicine,
Philadelphia, 2003.
be seen in the absence of a fixed coronary stenosis due
to small vessel disease or metabolic defects.
The reported sensitivity of stress myocardial perfusion
imaging has ranged from 70% to 95% and the specificity
from 50% to 90%. This wide range is due in part to differ-
ences in study populations. If patients with known mul-
tiple-vessel disease and prior myocardial infarction are
included in the study population, the sensitivity will be
predictably high. If only younger subjects with sus-
pected but not proven disease are studied, sensitivity will
be lower. Sensitivity for patients achieving adequate
exercise will be higher than if it is reported for all
patients studied.
Specificity ( negative imaging results in patients with-
out disease) is particularly difficult to determine.
Although causes for false-positive examinations for coro-
nary artery disease include cardiomyopathy, valvular
heart disease, and myocarditis, the problem of specificity
is more complex. In most institutions, the decision to
perform coronary angiography is based on the myocar-
dial perfusion scan. When a noninvasive test is accepted
as being clinically effective for diagnosis and risk stratifi-
cation, its results strongly influence whether coronary
angiography is done. If only patients with abnormal or
equivocal myocardial perfusion scans are sent for
482 NUCLEAR MEDICINE: THE REQUISITES
catheterization,the specificity of myocardial scintigraphy
in the “proven” population will be predictably low. Most
patients with normal studies will not have had the gold
standard test and thus the angiographic referral popula-
tion is very statistically biased.
Analysis of the results of many studies over the years
suggests that myocardial perfusion scintigraphy has sen-
sitivity and specificity in the range of 87% and 80%,
respectively.
Because true specificity cannot be determined, the
concept of “normalcy rate” has been developed. The
normalcy rate is determined in a group of patients
with a low likelihood of coronary artery disease, based
on Bayesian analysis using age, sex, symptom classifica-
tion, cholesterol, and results of noninvasive stress test-
ing. The normalcy rate is defined as the frequency of
normal test results in patients with a low likelihood of
coronary disease. Normalcy rates of greater than 90%
have been reported for SPECT myocardial perfusion
scintigraphy.
Other than observer and test performance, an interest-
ing and important observation in following patients over
time is that people with normal stress perfusion scans,
even if the results are false negatives based on anatomical
angiographic criteria, have a better prognosis than those
with scintigraphic evidence of ischemia. The myocardial
perfusion scan is a physiological test and the ultimate
gold standard is the outcome of the patient.
Prognosis and Risk Stratification
Although diagnosis continues to be an important indica-
tion for SPECT myocardial perfusion imaging,increasingly
common indications for stress myocardial perfusion
imaging are risk stratification and prognosis (Boxes 14-16
and 14-17).
Acute Ischemic Syndromes
Chest Pain in the Emergency Room More than
5 million patients present to the emergency room with
chest pain each year in the United States. Half are
admitted to the hospital. Ultimately only 5% are
diagnosed with myocardial infarction. Clinical decision-
making requires triage of patients into risk categories
based on the probability of infarct or unstable angina and
risk assessment ( Fig. 14-33). SPECT perfusion imaging
can provide information critical to this decision-making
process ( Fig. 14-34). The accuracy of diagnosis of acute
ischemic syndromes is highest when the radiopharma-
ceutical is injected during pain, although good accuracy is
obtainable for several hours thereafter.
Tl-201 is not used to diagnose acute ischemic syn-
dromes because the radiopharmaceutical requires imag-
ing promptly after injection. Tc-99m sestamibi or
tetrofosmin can be injected in the emergency room and
the patient transferred for imaging when initial evaluation
and stabilization is complete. Because the radiopharma-
Box 14-16 Prognosis and Risk
Stratification: Indications
Acute ischemic syndromes
Chest pain in the emergency room
Myocardial infarction
Unstable angina
Chronic ischemic syndromes
Known or suspected coronary artery disease
Assessment of coronary bypass grafts and angioplasty
Assessment of thrombolytic therapy
Assessment of percutaneous coronary intervention
Box 14-17 Indicators of Adverse
Outcome and Prognosis:
SPECT Perfusion Imaging
Increased lung to heart ratio after stress
Transient left ventricular cavity dilatation after exercise
Multiple and large reversible defects
Multiple and large irreversible defects
Reversible perfusion defects at low level exercise
ceutical is fixed and does not redistribute, delayed imag-
ing reflects blood flow at the time of injection. Negative
SPECT studies are highly predictive of a good prognosis.
Cardiac events occur in less than 1.5% of patients com-
pared to a 70% incidence in those with a positive study.
SPECT has a higher sensitivity than serum enzymes or
markers (e.g.,troponin). Positive enzymes or markers are
very specific, but require serial determination.
SPECT perfusion scintigraphy has a high sensitivity
(>90%) for detection of transmural infarction immedi-
ately after the event. Diminished or absent uptake is also
seen in the region of peri-infarct ischemia and edema.
However, the high sensitivity decreases with time as the
edema and ischemia resolve. By 24 hours after the acute
event, small infarctions may not be detectable and the
overall sensitivity for larger ones decreases. The sensitiv-
ity for detection is also less for nontransmural infarctions.
Furthermore, an acute infarction cannot be distinguished
from an old infarct.
Acute Myocardial Infarction with ST Elevation
Infarct size, left ventricular ejection fraction ( LVEF), and
residual myocardium at risk provide important prog-
nostic management information. Submaximal exercise
(achieving less than target heart rate) SPECT perfusion
scintigraphy postinfarction can detect the presence

Myocardial infarction suspected
History and physical examination,
ECG, enzymes, other laboratory studies
Infarction
established
hr
12
s

Therapy
S
Perfusion
scintigraphy
(Assessment of infarct
size and degree of
functional impairment)
Figure 14-33 Triage of acute chest pain. Simplified schematic of potential roles for scintigraphic
imaging and suspected myocardial infarction. S × s, symptoms.
and extent of stress-induced myocardial ischemia.
Pharmacologic SPECT scintigraphy with adenosine or
dipyridamole can be safely done as early as 2 to 5 days
postinfarction, earlier than the limited stress study.
Evidence for ischemia warrants aggressive management
(e.g., coronary angiography and revascularization). If
negative,the patient can be treated conservatively.
Unstable Angina and Non-ST Elevation Myocardial
Infarction Early invasive interventional therapy is
recommended for patients with high-risk indicators (e.g.,
positive myocardial perfusion scintigraphy). SPECT is used
for the predischarge risk stratification of patients with
unstable angina. Ischemia is seen in a high percentage
(90%) of patients who develop subsequent cardiac events
compared to those who do not have evidence of ischemia
(20%).
Chronic Ischemic Syndromes
Patient Management It is important to identify
patients at high risk but with minimal symptoms
whose mortality rate can be improved by coronary
bypass graft surgery ( Fig. 14-35). Low risk is defined as
less than 1% cardiac mortality rate per year versus high
risk with more than a 3% mortality. Many factors
assessed by SPECT determine patient prognosis,
including the extent of infarcted myocardium, the
amount of jeopardized myocardium supplied by
vessels with hemodynamically significant stenosis, and
the severity of ischemia.
Cardiac System 483
Infarction uncertain Unstable angina,
recurrent chest pain
S

s

12
hr
TI-201 rest and
redistribution
Tc-99m PYP scintigraphy
scintigraphy
A normal stress SPECT perfusion study predicts a good
prognosis,with less than a 1% annual risk of cardiac death
or infarct. With abnormal SPECT,the risk of cardiac death
or infarction increases. Poststress lung uptake and tran-
sient ischemic dilation suggest severe proximal LAD or
multivessel disease. These findings are associated with
high risk. Reduced LVEF is the strongest negative prog-
nostic predictor.
Assessment of Coronary Bypass Surgery and
Angioplasty Because 40–60% of angiographically
detected stenotic lesions are of uncertain significance,
myocardial perfusion scintigraphy can stratify risk and
assess which patients require revascularization. Those
patients with no ischemia have low risk for cardiac
events, even with left main or three-vessel disease on
angiography. Successful intervention results in the elimi-
nation of transient defects caused by exercise-induced
ischemia. Perfusion imaging should not be done before
6 weeks after intervention because some preinterven-
tion defects may persist.
Assessment of Thrombolytic Therapy Tc-99m
labeled perfusion agents can be given when a patient
arrives at the hospital and SPECT performed after the
patient is stabilized or after thrombolytic therapy.
The initial resting study documents the amount of
myocardium at risk. A second dose can then determine
the effectiveness of therapy. Reduction in defect size
correlates with vessel patency and improved prognosis
484 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-34 Emergency room chest pain: myocardial infarction. Resting SPECT study with Tc-
99m sestamibi. Radiopharmaceutical injection given in the emergency room and imaging delayed
until the patient was stabilized.The large defect involving the inferior wall of the heart is diagnostic
of infarction.

Post-MI evaluation of stable

chest pain syndromes

Perfusion scintigraphy
(stress and redistribution)

Single ‘‘fixed’’ defect, Fixed defect Multiple defects

no ECG evidence of with contiguous and/or defects in
ischemia reversible defect remote vascular zones

‘‘Completed infarction’’ ‘‘Incomplete infarction’’ Multivessel

disease

Conventional
medical therapy

Consider for further workup

(coronary angiography)
Figure 14-35 Risk stratification in chronic chest pain. Diagnostic scheme illustrating the
incorporation of perfusion scintigraphy into an approach for stratifying risk after myocardial
infarction. MI, Myocardial infarction; ECG, electrocardiogram.

after myocardial infarction. After recovery, stress per-
fusion scintigraphy can confirm therapeutic effectiveness
or detect areas of residual ischemia.
After Percutaneous Coronary Intervention
Symptom status and exercise ECG are unreliable
indicators of restenosis. Of patients with recurrent chest
pain within a month of intervention, 30% have restenosis.
Because ischemia (painful or silent) worsens prognosis,
stress SPECT perfusion may be useful.
After Coronary Bypass Surgery Abnormal per-
fusion patterns may reflect bypass graft disease, disease
in the native coronary arteries beyond the distal anasto-
mosis, nonrevascularized coronaries or side branches, or
new disease. SPECT perfusion scintigraphy can deter-
mine the location and severity of ischemia and has prog-
nostic value early and late after coronary bypass surgery.
When ischemia occurs 1–12 months after surgery, the
etiology is usually perianastomotic graft stenosis.
Ischemia developing more than 1 year postoperatively is
usually caused by new stenoses in graft conduits and/or
native vessels. Exercise SPECT is strongly predictive of
events.
Heart Failure: Assessment for Coronary Artery
Disease Heart failure in the adult can be due to vari-
ous etiologies, including hypertrophic cardiomyopathy,
hypertensive or valvular heart disease, and idiopathic
cardiomyopathy. Determining whether left ventricular
dysfunction is due to the consequences of coronary artery
disease or to other etiologies is critical for patient
management. If due to coronary disease,revascularization
can potentially reverse the left ventricular dysfunction.
Left ventricular dysfunction due to ischemic car-
diomyopathy is the result of either a large or multiple
prior myocardial infarctions with subsequent remodel-
ing, or moderate infarction associated with considerable
inducible ischemia and/or hibernation. The sensitivity
of SPECT myocardial perfusion scintigraphy approaches
100% for detection of coronary disease in patients with
cardiomyopathy; however, the specificity is only 50%.
False-positive studies are due to perfusion abnormalities
seen in many patients with nonischemic cardiomyopathy
(i.e., without epicardial coronary artery disease). Some
have regions of fibrosis and decreased coronary blood
flow reserve, resulting in both fixed and reversible
defects. More extensive and severe perfusion defects are
likely to be due to coronary artery disease while smaller
and milder defects are likely to occur in patients with
nonischemic cardiomyopathy.
Calcium Screening Electron-beam CT and multislice
helical CT measure the amount of calcium in the coronary
arteries (calcium score). The risk of cardiac events is low
with coronary calcium scores of 100 or less. Stress SPECT
myocardial perfusion tests are positive in less than 1% of
these patients. With scores between 101 and 399, risk of
Cardiac System 485
future cardiac events is moderate, with 12% having
abnormal stress SPECT perfusion studies. Scores greater
than 400 identify patients at high risk. Approximately 50%
of patients have abnormal SPECT studies.
Sarcoidosis
The clinical features of myocardial involvement of the
heart with sarcoidosis include dysrhythmias, conduction
defects, heart failure, and sudden death. Pathologically
any region of the heart can become the site of granuloma
deposition. Because of the serious potential conse-
quences of cardiac involvement, immunosuppressive
therapy with corticosteroids is indicated. However, the
diagnosis of cardiac involvement can be difficult.
Endometrial biopsy is confirmative; however, it is an
insensitive technique due to sampling error.
SPECT myocardial perfusion scintigraphy with Tl-201
and Tc-99m radiopharmaceuticals has been used to con-
firm or exclude myocardial involvement in patients with
sarcoidosis. Perfusion defects are common in the right and
the left ventricle and correlate with atrioventricular block,
heart failure, and ventricular tachycardia. One successful
approach has been to perform Persantine stress and rest Tc-
99m sestamibi studies. In most patients with cardiac sar-
coidosis, SPECT demonstrates a fixed defect. In some
patients, reverse “redistribution” is noted. Ga-67 has also
been used to diagnose cardiac sarcoidosis, but with lower
sensitivity. Preliminary data suggest F-18 FDG PET may be
useful to diagnose active cardiac sarcoidosis.
POSITRON EMISSION TOMOGRAPHY
OF THE HEART
Positron emission tomography ( PET) affords superior
spatial resolution compared to SPECT and routine attenu-
ation correction. The clinical use of PET for cardiac per-
fusion imaging is increasing, but still limited to certain
centers with high volume. The most common current
indication at most centers is the use of F-18 fluo-
rodeoxyglucose ( FDG) to diagnose hibernating ( viable)
myocardium. The use of rubidium ( Rb-82) as an alterna-
tive to Tl-201 and Tc-99m SPECT for perfusion imaging is
increasing because it is generator produced.
Radiopharmaceuticals for Myocardial
Perfusion
In addition to the advantage of better resolution and
superior attenuation correction with PET perfusion
agents over SPECT, multiple serial studies can be per-
formed within a brief period and absolute quantification
of coronary flow in milliliter/gram/minute is possible.
N-13 ammonia and Rb-82 are FDA-approved and reim-
bursable in the United States.
486 NUCLEAR MEDICINE: THE REQUISITES
Nitrogen-13 Ammonia
N-13 would be the preferred PET perfusion radiopharma-
ceutical except that it has a very short physical half-life
( 10 minutes), requiring onsite cyclotron production. It
decays 100% by positron ( β+) emission ( Table 14-7).
When injected intravenously, it clears rapidly from the
circulation with 85% leaving the blood in the first minute
and only 0.4% remaining after 3.3 minutes. Localization
in myocardial cells is from diffusion across the capillary
membrane, metabolic conversion to N-13-glutamine by
glutamine synthetase, and subsequent trapping within
tissues by incorporation into the cellular pool of amino
acids. Thus, the N-13 label remains within the heart with
a relatively long biological residence time. At physio-
logical pH, the major form of ammonia is NH4+. It has a
70–80% extraction rate by myocardial cells at normal
coronary flow rates.
Myocardial uptake is proportional to coronary blood
flow. As with other perfusion tracers, the extraction effi-
ciency of N-13 ammonia drops at higher flow rates. In
addition to the myocardium, it is taken up by the brain,
liver, and kidneys.
Technique
Ten to 20 mCi (370 MBq) of N-13 ammonia is adminis-
tered intravenously and imaging typically begins 5 min-
utes after injection, which allows time for pulmonary
background activity clearance. The long myocardial bio-
logical half-life offers some imaging timing flexibility. In
the diagnosis of coronary artery disease, a second study
is typically performed after pharmacologic stress (dipyri-
damole, adenosine, or dobutamine) with protocols simi-
lar to those described for SPECT myocardial perfusion
scintigraphy ( Fig. 14-36).
Radiation Dosimetry
Radiation absorbed dose to the patient is quite low
compared to most clinically used radiopharmaceuticals
( Table 14-8).
Rubidium-82 Chloride
Rb-82 is a generator produced positron radionuclide
(Table 14-7). In the strontium-82/Rb-82 generator system
Table 14-7 Cardiac Positron Radiopharmaceuticals
Mechanism Radionuclide Pharmaceutical
Perfusion Nitrogen-13 Ammonia
Rubidium-82 Rubidium
Oxygen-15 Water
Glucose metabolism Fluorine-18 Fluorodeoxyglucose
Fatty acid metabolism Carbon-11 Acetate
Carbon-11 Palmitate
( CardioGen-82, Bracco), the half-life of the Sr-82 parent is
25 days, which means that facilities using Rb-82 for
myocardial perfusion imaging need to receive one new
generator system each month. Thus, no onsite cyclotron
or pharmaceutical production facility is required.
However, the generator systems are expensive and a large
volume of cardiac studies is needed to make it financially
feasible. A commercial attempt to increase its clinical use
provides a generator that is transported from one PET site
to another. Rb-82 is increasingly being used on a clinical
basis.
Rb-82 is a monovalent cation and true analog of potas-
sium. Like thallium-201, Rb-82 is taken up into the
myocardium by active transport through the Na+K-
ATPase pump. Its extraction is somewhat lower than
that of N-13 ammonia ( 60%). The relative myocardial
extraction and localization of Rb-82 are proportional to
blood flow.
The very short half-life of Rb-82 ( 76 seconds) allows
the performance of sequential studies before and after
pharmacological interventions used for diagnostic PET
myocardial perfusion scintigraphy. Rb-82 decays 95% by
positron emission and 5% by electron capture. In addi-
tion to the beta particle + annihilation photons ( 511
keV), it emits a 776-keV gamma ( 15% abundance) and
1395-keV gamma (0.5% abundance).
Technique
Rb-82 (40–60 mCi) is infused intravenously over 30 to
60 seconds. Imaging is delayed for approximately 2
minutes to allow time for blood pool clearance; it is
completed within 5 minutes. Imaging time is short
because of rapid decay and to prevent reconstruction
artifacts. About 80% of the useful counts are acquired
in the first 3 minutes and 95% in the first 5 minutes.
Sequential studies can be performed within 10 minutes.
Rb-82 has the poorest resolution of positron radionu-
clides because the positrons travel about 13 mm prior
to undergoing annihilation. However, image quality is
superior to Tl-201 and it does not have the problems of
the Tc-99 radiopharmaceuticals (e.g., attenuation and
subdiaphragmatic scatter).
Physical half-life Production
10 min Cyclotron
76 seconds Generator
110 seconds Cyclotron
110 minutes Cyclotron
20 minutes Cyclotron
20 minutes Cyclotron
 Cardiac System 487

Figure 14-36 Importance of glucose loading. N-13 ammonia images (top row) and two sets of
F-18 FDG images of a diabetic subject.The N-13 ammonia images show a large perfusion defect at
the cardiac apex.The initial F-18 FDG images show essentially no myocardial uptake (middle row).
After insulin administration, FDG accumulates in the myocardium and reveals a matched defect at
the apex.

myocardium and it is not affected by metabolic factors.

Table 14-8 Radiation Dosimetry for Fluorine-18 Unlike other perfusion agents, extraction remains linear at
FDG, Rubidium-82, N-13 Ammonia very high flow rates, and therefore myocardial distribution
reflects regional perfusion. However, image quality is not
Fluorine-18 Rubidum- N-13 as good as that of other PET perfusion agents. Tracer circu-
FDG* 82† Ammonia‡ lating in the blood pool remains within the ventricular
chamber and must be subtracted in order to visualize the
Dose cGy/370 MBq cGy/1480 MBq cGy/740 MBq
(rads/ (rads/ (rads/ myocardium. With a half-life of 2.2 minutes (123 seconds),
10 mCi) 40 mCi) 20mCi) O-15 requires an onsite cyclotron for production.
Heart wall 2.5 0.4 0.2
Liver 0.8 0.16 0.3
Kidneys 0.8 2.8 0.2 Diagnosis of Coronary Artery Disease
Brain 1.7 0.3
Urinary 3.0 The two PET perfusion agents, Rb-82 and N-13 ammonia,
bladder are used clinically for the diagnosis of coronary artery
Colon 0.4 disease. Cardiac PET stress tests use pharmacologic
Thyroid 0.4 5.6 0.2 stress rather than exercise because of the short half-life
Ovaries 0.4 0.04 0.2
of the radiopharmaceuticals. After baseline studies are
Red marrow 0.16 0.2
Effective dose 1.0 0.8 0.2 obtained under resting conditions, one of the pharmaco-
logical agents is administered to challenge coronary flow
Target organ (highest radiation absorbed dose) in boldface type. reserve. The protocols are the same as in single-photon
*, MIRD; †, IRCP; ‡, IRCP. imaging. The timing of injection of the PET radiophar-
maceutical is synchronized with the administration of
Dosimetry the pharmacological agent and the desired delay after
The thyroid receives the highest radiation dose (5.6 cGy/ injection before the onset of imaging.
1480 MBq or 5.6 rads/40 mCi) ( Table 14-8), followed by PET imaging offers superior spatial image resolution
the kidney (2.8 cGy/1480 MBq or 2.8 rads/40 mCi). compared to SPECT and typically excellent target-to-
background ratios. The scintigraphic appearance of the
Oxygen-15 Water heart and diagnostic criteria for Rb-82 and N-13 ammo-
O-15 water is ideal for quantitative regional myocardial nia studies are the same as for perfusion scans obtained
flow measurements (ml/min/gm) because it is a freely dif- with Tl-201 or the Tc-99m-labeled perfusion agents.
fusible perfusion tracer with 95% extraction by the Normal subjects have homogeneous uptake of the tracer
488 NUCLEAR MEDICINE: THE REQUISITES
throughout both the left and right ventricular myocar-
dium. Patients with hemodynamically significant CAD but
no ischemia at rest demonstrate normal resting myocar-
dium. Perfusion defects are seen after pharmacological
stress. Areas of prior myocardial infarction appear cold on
both baseline and poststress images. Patients with hiber-
nating myocardium also demonstrate persistent perfusion
defects or both phases. The diagnostic patterns of
ischemia and infarction are similar to SPECT (Table 14-6).
The sensitivity of PET for the diagnosis of coronary dis-
ease is high, on the order of 95%. The specificity reported
in the early literature is also high. The specificity should
be regarded with caution because early reports under clin-
ical research protocols frequently use normal volunteers
to determine specificity, which is very different from
determining specificity in a more broadly chosen cross
section of patients with and without coronary artery dis-
ease. Furthermore, there is much more limited data on
the accuracy of PET than SPECT. Generally the specificity
of PET is felt to be superior to SPECT because of the stan-
dard direct method of attenuation correction. Breast and
subdiaphragmatic attenuation artifacts are much less of
a problem with high-energy PET radiopharmaceuticals
than with SPECT.
PET Metabolic Radiopharmaceutical
Fluorine-18 Fluorodeoxyglucose
F-18 FDG is a marker of myocardial glucose metabolism.
The physical half-life of F-18 is 1.8 hours (110 minutes)
and thus must be delivered from a regional cyclotron on
a daily or twice daily basis. Only about 1–4% of the
injected dose is trapped in the myocardium; however
there is a high target-to-background ratio. Blood clear-
ance of FDG is multicompartmental and takes much
longer than the perfusion agents. Imaging is typically
begun 45 to 60 minutes after tracer injection (10–15 mCi
[370–555 MBq]) to allow maximal myocardial uptake
and blood and soft tissue background clearance.
F-18 has the best resolution of all positron emitters,
approaching 2 mm because most of the emitted positrons
travel only about 1.2 mm prior to undergoing annihilation.
Dosimetry
The organ receiving the highest radiation absorbed dose is
the urinary bladder (3.0 cGy/1480 MBq or 3.0 rads/40 mCi)
(Table 14-8).
Clinical Indication
Its principal use in practice is in combination with a per-
fusion tracer to assess myocardial viability. Under normal
conditions, most of the energy needs of the heart are met
through fatty acid metabolism. However, areas of
ischemia switch preferentially to glucose metabolism and
have increased uptake of F-18 FDG relative to perfusion.
Regional myocardial uptake of F-18 FDG reflects regional
rates of glucose utilization. In the myocardial cell, FDG is
phosphorylated to FDG-6-phosphate. No further metabo-
lism takes place, and the radiopharmaceutical stays
trapped in the myocardial cell over a prolonged period.
The state of glucose metabolism in the body highly
influences the amount of FDG taken up in the heart.
Myocardial glucose utilization is increased by glucose
administration which stimulates insulin secretion. The
increased insulin levels stimulate glucose metabolism.
Thus high serum glucose and insulin levels and low free
fatty acids promote uptake ( Fig. 14-36).
Different strategies are used to promote optimal FDG
uptake. Glucose loading with either oral glucose (glu-
cola) or intravenous dextrose is the most common
method ( Box 14-18). Serum glucose is checked
to ensure euglycemia. F-18 FDG is usually injected
45–60 minutes later. Diabetics often have an attenuated
increase in plasma insulin levels following glucose load-
ing and small intravenous insulin doses are required.
A more sophisticated method to ensure euglycemia,
used primarily in research settings, is hyperinsulinemic
euglycemic clamping. Glucose is infused in one arm
and insulin in another. The rate is varied to optimize to
euglycemia. In all three methods, F-18 FDG is infused
when euglycemia is obtained, and imaging is initiated
45 minutes later.
Detection of Myocardial Viability
The combination of perfusion imaging and metabolic
imaging with FDG can be of great benefit in correctly
diagnosing and assessing the potential therapeutic out-
come in patients with severely ischemic or hibernating
myocardium ( Table 14-9). The rationale for using FDG
is that severely ischemic myocardium switches from fatty
acid metabolism selectively to glucose metabolism.
Thus, FDG uptake can actually be greater in the ischemic
areas than in the remainder of the myocardium.
In normal subjects, perfusion and FDG uptake are
matched ( Fig. 14-37). The combination of matched per-
fusion and FDG scintigraphic defects is indicative of
myocardial scarring ( Fig. 14-38). The combination of
a photon-deficient area by perfusion imaging that
demonstrates increased FDG uptake is the scinti-
graphic hallmark of severely ischemic or hibernating
myocardium ( Fig. 14-39).
Perfusion/FDG mismatch indicates myocardial viabil-
ity. In areas of flow-metabolism mismatch, the functional
prognosis following revascularization is very good, with
an average of 80% of such segments demonstrating con-
tractile improvement after coronary artery bypass sur-
gery. The matched pattern of abnormal decreased
perfusion and glucose metabolism indicates a low likeli-
hood of improved function after therapeutic interven-
tion with only an average of 15% of such segments
demonstrating contractile improvement after coronary
artery bypass grafting.
 Cardiac System 489

Box 14-18 Protocol: F-18 FDG PET Cardiac Viability

PATIENT PREPARATION
NPO after midnight
Obtain rest myocardial perfusion scan
Serum fasting blood sugar (BS)
Nondiabetic
If BS ≤ 150 mg/dL: 50 gm oral glucose solution + regular insulin 3 units IV
If BS 151–300 mg/dL: 25 gm oral glucose solution + regular insulin 3 units IV
If BS 301–400 mg/dL: 25 gm oral glucose solution + regular insulin 5 units IV
If BS > 400 mg/dL: 25 gm oral glucose solution + regular insulin 7 units IV
At least 45 minutes after glucose loading and when BS ≤150 mg/dL, inject F-18 FDG
Diabetic
If BS ≤150 mg/dL: 25 gm oral glucose solution
If BS 151–200 mg/dL: Regular insulin 3 units IV
If BS 201–300 mg/dL: Regular insulin 5 units IV
If BS 301–400 mg/dL: Regular insulin 7 units IV
If BS 401 mg/dL or greater: Regular insulin 10 units IV
Obtain BS every 15 minutes for 60 minutes. If BS elevated, additional insulin per scale. At least 45 min after glucose
loading and when BS ≤150 mg/dL, inject F-18 FDG

DOSAGE AND ROUTE OF ADMINISTRATION

F-18 fluorodeoxyglucose 0.22 mCi/kg (100 μCi/lb)

TIME OF IMAGING
After 60 minute uptake phase

PROCEDURE
PET acquisition—cardiac field of view

PROCESSING
Reconstruct along the short and long axis of the heart similar to the perfusion study

IV, Intravenously; BS, serum blood glucose.

has been investigated and shows considerable promise.

Table 14-9 PET Diagnostic Image Patterns: The drug, administered orally, lowers free fatty acid levels
Perfusion and Metabolic Imaging by inhibition of lipolysis and thus enhances cardiac glucose
and F-18 FDG uptake. Preliminary studies are encourag-
Glucose ing. Further investigations are necessary.
Perfusion (N-13 metabolism
Diagnosis ammonia, Rb-82) (FDG)
Combined Tc-99m Sestamibi and F-18 FDG
Normal myocardium Present Present Imaging
Ischemic myocardium Absent or decreased Present SPECT-PET cardiac studies can be acquired using specially
Myocardial infarction Absent Absent modified gamma scintillation cameras either with high-
energy collimators and appropriately shielded camera
Patients with viable but severely ischemic myocar- heads or, alternatively, dual-headed cameras with coinci-
dium have better survival and event outcomes from sur- dence circuitry without a collimator. With the use of two
gical revascularization than from medical management. energy windows, a simultaneous combination of Tc-99m
Patients with only myocardial scarring and no ischemia sestamibi or tetrofosmin perfusion imaging and F-18 FDG
do not benefit from revascularization surgery. Some may metabolic imaging is possible.
be candidates for cardiac transplantation. An advantage of doing simultaneous imaging is that
Acipimox the data from the two radiopharmaceuticals are perfectly
Glucose loading and insulin administration can be time- registered, allowing optimal comparison of the respec-
consuming and not always effective, particularly in diabet- tive uptake patterns. This elegant approach uses the
ics. As an alternative, acipimox, a nicotinic acid derivative, same diagnostic criteria described previously. Areas of
490 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-37 F-18 FDG and N-13 ammonia PET in a normal Figure 14-39 F-18 FDG and N-13 ammonia: match. Decreased
subject.The uniform uptake of both tracers is concordant with a uptake in the region of the septum and apex.The concordant
normal appearance of the heart. pattern of matched abnormalities indicates myocardial scar with
absence of perfusion and metabolism.

matched perfusion and metabolic abnormality are

unlikely to improve ( Fig. 14-40). Areas demonstrating
diminished Tc-99m activity with normal or increased
FDG represent ischemic but viable tissue and have a high
likelihood of functional recovery after revascularization
(Fig. 14-41).

RADIONUCLIDE VENTRICULOGRAPHY

Radionuclide ventriculography ( RVG) or multigated

Figure 14-38 F-18 FDG and N-13 ammonia: mismatch. On the
N-13 ammonia perfusion study, uptake is decreased anteroapically.
The same area demonstrates good uptake by FDG.This discordant
pattern indicates diminished perfusion to an area of viable
myocardium. Prognosis for functional improvement after coronary
artery bypass grafting is good with this pattern.
acquisition ( MUGA) has been used since the 1970s to
evaluate global and regional right and left ventricular
function. Two different methods have been used, first-
pass studies, in which all data collection occurs during
the initial transit of a tracer bolus through the central cir-
culation, and equilibrium studies, in which data are col-

Tc
Short
axis

FDG

Tc
long axis
Vertical

FDG

Tc
Horizontal
long axis

FDG

Figure 14-40 Combined Tc-99m sestamibi and F-18 FDG imaging: mismatch.Tc-99m perfusion
imaging shows a large inferoapical defect that corresponds with normal uptake of FDG.This
discordant pattern is indicative of ischemic but viable myocardium.
 Cardiac System 491

Tc

Short
axis
FDG

long axis Tc
Vertical

FDG

Tc
Horizontal
long axis

FDG

Figure 14-41 Tc-99m perfusion imaging and F-18 FDG metabolic imaging. Simultaneously
acquired study reveals matched abnormalities in the inferior wall of the left ventricle.This pattern of
matched abnormalities is indicative of myocardial nonviability.

lected over many cardiac cycles using ECG gating and

a tracer that remains in the blood pool. RVG can be suc- Box 14-19 Causes of Poor Tc-99m Red
cessfully performed in most all patients and ejection frac- Blood Cell Labeling
tion quantification is not dependent on mathematical
assumptions of ventricular shape, as in contrast ventricu- Drug-drug interactions Heparin, doxorubicin,
lography and echocardiography. methyldopa, hydralazine,
contrast media, quinidine
Circulating antibodies Prior transfusion, transplan-
Radiopharmaceuticals tation, some antibiotics
Blood Pool Agents Too little stannous ion Insufficient to reduce
Tc (VII)
The radiopharmaceutical of choice for equilibrium
Too much stannous ion Reduction of Tc (VII) outside
gated blood pool imaging is Tc-99m-labeled red blood
of red blood cell before
cells ( Tc-99m RBCs). Labeling may be accomplished by cell labeling
any of three approaches: in vivo, modified in vivo, and Carrier Tc-99 Buildup of Tc-99m in the
in vitro. These methods are described in detail in Mo-99/Tc-99m generator
Chapter 11 ( Box 11-13). Causes for poor Tc-99m red due to long interval
blood cell labeling are listed in Box 14-19. The modified between elutions
in vivo or the in vitro method (Ultratag RBC,Mallinckrodt) Too short an interval Not enough time for
are preferred because of their higher binding efficiency for “tinning” stannous ion to penetrate
(85–90% for the modified in vivo method, 98% for the the red blood cells before
in vitro method). addition of Tc-99m
Too short an incubation Not enough time for
Dosimetry
time reduction of Tc (VII)
The spleen receives the highest radiation absorbed dose,
2.2 cGy/740 mBq ( 2.2 rads/20 mCi), followed by the
heart wall. 2.0 cGy/740 mBq (2.0 rads/20 mCi). See
Chapter 11 ( Table 11-6).
is its high residual background activity if multiple studies
First-Pass Agents are required. Tc-99m DTPA is alternatively used because
Any Tc-99m labeled agent may be administered as a bolus of its rapid renal excretion. Using Tc-99m RBCs allows
for first-pass imaging of the central circulation. Tc-99m as combining a first pass evaluation of the right ventricle and
sodium pertechnetate is commonly used. A disadvantage equilibrium analysis of the left ventricle. Some institutions
492 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-42 First-pass radionuclide angiogram. Cardiac structures are sequentially visualized as
the bolus passes through the right side of the heart into the lungs and then returns to the left side.
Ao, Aorta; AV, aortic valve; LA, left atrium; Lu, lung; LV, left ventricle; PA, pulmonary artery; RA, right
atrium; RV, right ventricle; SVC, superior vena cava; TV, tricuspid valve.

perform a first-pass study with one of the Tc-99m-labeled
myocardial perfusion agents in conjunction with the per-
fusion study.
Acquisition Techniques
First-Pass Studies
First-pass studies are obtained by injecting a compact
bolus of the selected radiopharmaceutical intravenously
preferably via the jugular vein. If a peripheral injection
is used, the Oldendorf technique or a variation thereof is
employed. The arm is held in a neutral position, and
a medial vein in the basilic system is used at the antecu-
bital fossa. Use of veins in the cephalic system should
be avoided, if possible, to prevent “hang-up” of the
bolus at the thoracic inlet. Injections directly through
central catheters placed in the superior vena cava pro-
vide the most compact boluses. Jugular venous access
is an alternative effective approach.
Data may be acquired either in rapid frame mode or
in list mode, with or without ECG gating. Whichever
approach is used, the goal is to obtain 16–30 frames
per second while the bolus passes through the central
circulation. In most patients, the total data acquisition
time required is on the order of 30–60 seconds or less.
Typically a right anterior oblique view at 20- to 30-
degree angulation is chosen ( Fig. 14-42). This view
best separates the right atrium and the right ventricle
and is also one of the standard views of the left ventri-
cle used during cardiac catheterization. It is suitable
for both quantitative and qualitative analysis of biven-
tricular function.
The major advantage of the first-pass approach is that
data are collected rapidly over very few cardiac cycles.
Ventricular function can be measured at peak stress dur-
ing exercise ventriculography or other intervention.
Right ventricular function quantification is more accurate
than with equilibrium gated blood pool studies, in which
there is overlap of the right and left ventricles in the RAO
view and between the right atrium and the right ventricle.
The major disadvantage of the first-pass or first-transit
approach is that counting statistics are low in each frame
because of the count rate limitations of gamma cameras.
Special multicrystal gamma cameras with their high
count rate capability are optimally suited for first-pass
studies but are not widely available. In current practice,
equilibrium gated blood pool studies are performed
much more frequently than first transit studies.
Equilibrium Gated Blood Pool Studies
The limited counting statistics available during any one
cardiac cycle and the desirability of linking phases of the
cardiac cycle to image data underlie the equilibrium gated
blood pool approach to RVG. In this approach, ECG leads
are placed on the patient and a gating signal that triggers
the R wave of the ECG is sent to the nuclear medicine
computer system ( Fig. 14-43). The R wave is a useful
marker because it occurs at the end of diastole and the
beginning of systole. It is the largest electrical signal in
the normal ECG and therefore relatively easy to detect.
The cardiac cycle is divided into 16 frames in most
commercially available computer systems ( Fig. 14-44).
Individual frame duration is 40–50 msec. This frame rate
is a compromise between optimal temporal and statistical
 Cardiac System 493

X, Y, Z pulses Gamma (X, Y) Nuclear medicine

camera minicomputer
console
ma
Gam era
cam
ECG synchronizer R Wave
timing pulse
Data
acquisition,
Nonfade storage,
oscilloscope analysis,
display
Hardcopy
record of ECG

Figure 14-43 Acquisition of R-wave gated radionuclide ventriculography.A special ECG

synchronizer or gating device depicts the R-wave and sends a timing pulse to the nuclear medicine
computer system.This timing pulse is used to sort incoming scintillation events into a sequence of
frames that spans the cardiac cycle.

Figure 14-44 Sequential frames of R-wave gated MUGA study.Anterior (A) and the 45º left
anterior oblique views (B). In this study, the cardiac cycle was divided into 16 frames. Note the
change in size and count density of the cardiac chambers through the cardiac cycle.
Continued
494 NUCLEAR MEDICINE: THE REQUISITES
Figure 14-44, cont'd
data sampling. Enough frames are needed to catch the
peaks and valleys of the cardiac cycle ( temporal sam-
pling), but too many frames reduce counting statistics
available in any single frame (statistical sampling).
During each heartbeat, data are acquired sequentially
into the frame buffers spanning the cardiac cycle. With
imaging of more than 100–300 cardiac cycles, sufficient
counting statistics are obtained for valid quantitative
analysis and reasonable spatial resolution. Studies at rest
are obtained for 250,000 counts per frame. Studies
obtained during exercise or other intervention are often
obtained for somewhat fewer counts per frame to cap-
ture the peak effect of the stress. Box 14-20 describes
the equilibrium gated blood pool ventriculography pro-
tocol in detail.
The underlying assumption of R-wave gating is the
presence of normal sinus rhythm so that data are added
together from corresponding segments of the cardiac
cycle over the entire time of the study. Any significant
arrhythmia degrades the quality of the data and reduces
the accuracy of quantitative analysis.
A rhythm strip should be obtained for every patient
before the injection of a radioactive tracer to deter-
mine suitability for examination. For example, rapid
atrial fibrillation with an irregular ventricular response
or frequent premature ventricular contractions is
a contraindication to the study ( Fig. 14-45).
Quantitative error may result if there are greater than
10% premature ventricular contractions or a rapid
atrial fibrillation with irregular ventricular response.
Recording a beat histogram throughout the study is
useful ( Fig. 14-46). Problems with gating include spu-
rious signals from skeletal muscle activity, giant
T waves triggering the gating device, and artifacts from

Box 14-20 Equilibrium Gated Blood Pool
Ventriculography: Protocol
Summary
PATIENT PREPARATION AND PRECAUTIONS
Rhythm strip to confirm normal sinus rhythm (less than
10% premature ventricular contractions)
DOSAGE AND ROUTE OF TRACER ADMINISTRATION
Tc-99m red blood cells 740 MBq (20 mCi) intravenously
IMAGING PROTOCOL
Collimator: low-energy, general purpose collimator and a
15–20% window centered at 140 keV
Imaging: anterior, left anterior oblique (LAO) (best septal
view), and left lateral views.
Gamma camera persistence scope to determine optimal
LAO view for separating left and right ventricular
activity
Sixteen frames per cardiac with frame duration
of 50 msec or less
250k counts per frame for studies performed at rest
100k counts per frame in the optimum LAO view for
studies obtained during an intervention
pacemakers. The pacemaker signal itself is usually a
reliable trigger for gating.
Special computer techniques may be used to filter
data from premature contractions and postextrasystolic
beats, but these increase the time needed to perform
a study. By the same token, elegant gated list mode data
acquisition techniques have been developed to analyze
separately the normal sinus beat, the premature con-
traction, and the postextrasystolic beat. List mode is
not commonly used, in the past because of the large
computer memory required, but nowadays because of
the less frequent use of RVGs and the adequacy of equi-
librium studies.
For studies at rest, multiple views ( anterior, LAO, left
lateral) are obtained to provide the most comprehen-
sive evaluation of regional ventricular wall motion
(Fig. 14-47). The exact angulation for the LAO view is
determined empirically by moving the head of the
gamma camera to find the view that best separates
activity in the left and right ventricles for most accurate
calculation of the LVEF.
Protocols for exercise RVGs have varied among insti-
tutions. Treadmill and bicycle ergometer exercise have
been used. The LVEF can be determined for each stage
of a graded exercise program designed to recapitulate
graded treadmill stress, or more commonly a baseline
Cardiac System 495
study and a single stress study during peak exercise is
obtained. In addition to exercise stress, alternatives
have been used, including cold pressor testing, hand-
grip isometric exercise, atrial pacing, and pharmacologi-
cal stress. None of the alternatives have proved equal
to leg exercise studies. Because of the general use of
gated SPECT, exercise RVGs are performed uncom-
monly today.
Data Analysis and Study Interpretation
Qualitative Analysis
Comprehensive analysis and interpretation of RVGs
require both qualitative and quantitative assessments
( Box 14-21). Wall motion is analyzed by viewing a repet-
itive cinematic closed loop display on the computer
screen. Ventricular contraction is inferred from “shrink-
age” of the ventricular activity from diastole to systole.
Failure of activity to diminish or clear along the ventricu-
lar periphery is an indication of abnormal wall motion.
Septal contraction is inferred from seeing the photon-
deficient area between the right ventricular and left ven-
tricular blood pools thicken during systole.
Complete absence of wall motion is termed akinesis.
Areas with diminished contraction are hypokinetic.
Those demonstrating paradoxical wall motion ( that is, an
actual outward bulge during systole) are termed dyski-
netic. If motion is still present but delayed compared
with adjacent segments, it is referred to as tardokinesis.
In normal subjects, all wall segments should contract,
with the greatest incursion seen in the left ventricular
free wall and apex. Areas of ventricular scar are typically
akinetic or dyskinetic. Areas of ventricular ischemia
become hypokinetic with exercise. Tardokinesis is seen
with ischemia or conduction abnormalities such as
bundle-branch block.
The complete qualitative or visual analysis includes
an assessment of cardiac chamber size for all four car-
diac chambers, assessment of overall biventricular func-
tion and regional wall motion, and assessment of any
extracardiac abnormalities such as aortic aneurysms or
pericardial effusions that are in the detector’s field of
view. Only portions of the ventricles not overlapped by
other cardiac structures should be assessed on any given
view. For example, on the anterior view the right ventri-
cle usually overlaps the septum and inferior wall of the
left ventricle.
Attempts have been made to use quantitative and func-
tional or parametric images to detect abnormalities in
regional wall motion. Regions of interest may be flagged
along the ventricular perimeter to calculate regional ejec-
tion fractions. Fourier phase analysis and other paramet-
ric image analysis techniques are described in the
following section.
496 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-45 Electrocardiographic rhythm strips. Patients referred for gated radionuclide
ventriculography. A, Normal sinus rhythm. B, Sinus tachycardia. C, Atrial fibrillation with irregular
ventricular response. D, Ventricular premature contractions with bigeminy.

Figure 14-46 Beat histogram.The number of recorded beats for each observed cardiac cycle
length are depicted.This demonstrates that the heart rate and therefore beat length varied
significantly during data collection.

Figure 14-47 Anatomy at end-systole and end-diastole;
correlation with radionuclide ventriculogram A, End-diastolic
images from a gated radionuclide ventriculogram.Anterior (top
left), left anterior oblique (top right ), and left posterior oblique
(bottom) views are the most commonly obtained. B, Drawings
over left anterior oblique end-diastolic (left) and end-systolic
(right) frames, indicating position and relationships of major
structures.
Quantitative Data Analysis
Ejection Fraction
The most frequently calculated quantitative parameter of
ventricular function is the LVEF, defined as the fraction of
the left ventricular end-diastolic volume expelled during
contraction. The underlying principle is that the left ven-
tricular count rate at each point in the cardiac cycle is
proportional to ventricular volume. The ventricular
counts are determined by drawing a region of interest
over the left ventricle for each frame ( Fig. 14-48) of the
Cardiac System 497
Box 14-21 Diagnostic Criteria for
Exercise Radionuclide
Ventriculography
NORMAL BASELINE
Resting ejection fraction >50%
No regional wall motion abnormalities
Normal ventricular chamber size
NORMAL RESPONSE TO EXERCISE
Ejection fraction increases by 5% with continued
normal regional wall motion
ABNORMAL RESPONSE TO EXERCISE
Failure of ejection fraction to increase or decrease and/
or development of a new wall motion abnormality
cardiac cycle and a background region, typically taken as
a crescent adjacent to the left ventricular apex ( Fig. 14-
48). A background-corrected ventricular time–activity
curve is generated ( Fig. 14-49). End-diastole is the frame
demonstrating the highest counts and end-systole the
frame with the fewest counts.
Ejection fraction is calculated as follows:
End diastolic count − End systolic count
Ejection fraction =
End diastolic count
The LVEF in normal subjects ranges from 55–75%.
Many use 50% as a cutoff for normal. The accuracy of the
LVEF calculation by RVG is considered very good,superior
to that of nonnuclear techniques such as echocardiogra-
phy. Numerous studies have demonstrated good correla-
tion with contrast-enhanced left ventriculography. Most
consider the RVG calculation to be the most accurate
method.
The time–activity curve should be inspected in each
case as a quality control measure. Theoretically, the count
values at the beginning and end of the curve should be
identical. In practice, the trailing frames in late diastole
usually have fewer counts,owing to slight variations in the
length of the cardiac cycle, even in patients with normal
sinus rhythm (Figs. 14-49 and 14-50). In patients with fre-
quent PVCs, the falloff in counts at the end of the curve is
much greater. In atrial fibrillation with an irregular ven-
tricular response, a marked falloff may occur because car-
diac cycles of widely varying length are being added
together. Quantitative analysis of gated data in cases with
major dysrhythmias is not accurate. On cine display,
a falloff in counts in later frames is seen as a flicker.
Numerous other quantitative parameters have been
proposed for calculation from equilibrium gated blood
pool examinations ( Box 14-22). None of these can be
considered as well documented and validated as the
498 NUCLEAR MEDICINE: THE REQUISITES

Figure 14-48 Calculation of the left ventricular ejection fraction.A region of interest is defined
over the left ventricle in each frame of the cardiac cycle.A time–activity curve is then generated.The
percent LVEF is calculated utilizing the end-diastolic and end systolic ventricular count rates (ED − ES)
divided by (ED) × 100.

Figure 14-49 Composite computer-generated display: analysis of gated radionuclide

ventriculogram.The sequential left anterior oblique views are displayed across the bottom.The end-
diastolic and end-systolic regions of interest are indicated, along with the crescent-shaped
background region of interest adjacent to the left ventricle at end-systole (bottom row, second
image from right).The three parametric images in the upper right-hand corner represent ejection
fraction (ES – ED), paradox (ES – ED), and amplitude. The LVEF of 63% is normal.
 Cardiac System 499

Figure 14-50 Late frame count rate loss due to sinus arrhythmia. Small changes in beat length
result in fewer counts being recorded in the trailing frames of late diastole. Note how the last data
point in the volume curve is lower than the one adjacent to it and the trailing end of the volume
curve is somewhat lower than the beginning.

Calculation of the right ventricular ejection fraction

Box 14-22 Functional Parameters
Determined on Equilibrium
Blood Pool Ventriculograms
Wall motion assessment (regional and global)
End-diastolic and end-systolic ventricular volume
Stroke volume
Cardiac output
Ejection fraction (left and right ventricles)
Regurgitant fraction (stroke index ratio)
Ventricular filling and emptying rates (dV/dt) (peak and
average)
Cardiac shunt quantification
LVEF. Calculation of stroke volume and cardiac output
requires correction of the left ventricular count rate for
soft tissue attenuation, which is subject to considerable
error.
( RVEF) from equilibrium data has problems because of
overlap of chambers.
Fourier Phase Analysis
Fourier phase analysis reduces four-dimensional data into
a pair of two-dimensional images. These images portray
cardiac contractility ( amplitude) and contraction
sequence ( phase) ( Fig. 14-51). Simply stated, each pixel
in the cardiac image can be considered to have its own
cycle, having an amplitude and a characteristic temporal
relationship ( phase) with respect to the R wave. The
amplitude image simply portrays the maximum net count
variation for each pixel during the cardiac cycle. The
phase image portrays the relative time delay from the
R wave to the start of the cardiac cycle for that individual
pixel.
If the complete cardiac cycle is taken as encompass-
ing 360 degrees, the atria and ventricles are normally
180 degrees “out of phase” ( see Fig. 14-51). Areas of the
ventricle that contract slightly earlier in the cardiac
500 NUCLEAR MEDICINE: THE REQUISITES
Figure 14-51 Regional ejection fraction. Calculation of
regional ejection fractions from eight pie-shaped regions centered
in the middle of the left ventricle.The stroke volume, paradox, and
amplitude images are also illustrated.
cycle owing to the pattern of the electrical conduction
down the septum and through the bundle branches are
seen to be out of phase with adjacent ventricular
areas.
Wall motion abnormalities are portrayed on phase
images as low-amplitude areas. Regions of paradoxical
motion resulting from left ventricular aneurysms, for
example, are 180 degrees “out of phase” with the ventri-
cle. Abnormal conduction patterns like those seen in
Wolff-Parkinson-White syndrome or bundle-branch block
cause affected areas to be out of phase with adjacent
Figure 14-52 Phase and amplitude. In addition to the amplitude image that portrays cardiac
contractility, the phase parametric image portrays the relative time of contraction for each pixel on
the image.The atria and ventricles are normally 180 degrees “out of phase.”
portions of the ventricle owing to premature or delayed
contraction.
Amplitude and phase maps are often displayed in
color to highlight the temporal sequences of cardiac
chamber emptying. A dynamic color display mode can
be used to demonstrate the propagating wavefront that
sweeps across the ventricle during contraction, linking
pixels with similar phase angles together.
Although Fourier phase analysis is elegant, the studies
require exceptionally well-synchronized data to be useful
for localizing abnormal conduction pathways. Amplitude
and phase images are often presented automatically as part
of computer analysis packages and are useful for cueing the
observer to areas of abnormal wall motion.
Functional Images
The intensity of the computer display at each point in an
image is determined by the number of counts recorded
at that point and proportional to the amount of radioac-
tivity in the corresponding location. By subtracting the
end-systolic image from the end-diastolic image point by
point, a derived functional image is created that portrays
regional stroke volume. The stroke volume image may
be further processed by dividing it point-by-point by the
end-diastolic frame to create an “ejection fraction” image
( Fig. 14-52). In these images, akinetic wall segments cor-
respond to areas of diminished or absent intensity. In the
paradox image the end-diastolic frame is subtracted
from the end-systolic frame. With normal ventricular
function, this leaves a void. In patients with areas of

paradoxical ventricular wall motion, the systolic bulge is
readily detected as an area of unsubtracted activity.
A complete analysis and interpretation of the RVG
includes a qualitative visual assessment of the cardiac
chambers and great vessels to assess their size and rela-
tionships. Visual assessment of the dynamic cinematic
display is also used to analyze regional wall motion.
Quantitative analysis includes at a minimum calculation
of the LVEF. For specific applications other quantitative
parameters such as left and right ventricular stroke vol-
ume ratios, cardiac output, ventricular volume, and rates
of ventricular filling and emptying may also be calculated
but require more sophisticated analysis and in some
cases more sophisticated data acquisition techniques.
Clinical Applications
In the past, both myocardial perfusion scintigraphy and
the RVG were used for similar indications, depending on
personal preference and experience. Today, SPECT
myocardial perfusion scintigraphy has become the stan-
dard radionuclide study for evaluation of coronary artery
disease. RVG is reserved for specific indications. The
most common clinical indication is to evaluate ventricu-
lar function and calculate an ejection fraction in patients
receiving cardiotoxic drugs. This section emphasizes
that indication and others review current indications, as
well as a brief description of how it has been used in the
past and the associated scintigraphic findings that have
pertinence for specific and more general interpretation
of RVGs.
Coronary Artery Disease
Many patients with coronary artery disease have normal
resting ventricular function. Some will have regional
motion abnormalities ( Fig. 14-53). Exercise-induced
myocardial ischemia can be detected with RVG. The
hallmarks of ischemia are the development of a new wall
Figure 14-53 Resting apical hypokinesis due to myocardial
infarction. Selected end-diastolic and end-systolic images for a
patient with acute anteroapical myocardial infarction.The apex is
akinetic (arrow).The right ventricle and other portions of the left
ventricle show good contraction.
Cardiac System 501
motion abnormality during exercise stress testing that
was not present at rest and an ejection fraction that fails
to increase or even decreases in response to exercise
( Box 14-21, Fig. 14-54).
In patients able to achieve adequate levels of exercise,
the technique is highly sensitive ( on the order of 90%)
for the detection of coronary artery disease. However,
an abnormal ventricular functional response to exercise
stress is nonspecific. A low LVEF is seen in many other
cardiovascular diseases. Thus the specificity of the RVG
for coronary artery disease is not high and varies with
patient referral population ( Box 14-23). In current prac-
tice, myocardial perfusion scintigraphy with gated SPECT
is used to diagnose coronary artery disease.
The hallmark of an acute myocardial infarction on the
RVG is that of a wall motion abnormality in the region of
the infarct ( see Fig. 14-53) and a reduced regional and
global LVEF. Patient prognosis after acute myocardial
infarction is directly linked to the degree of functional
impairment. Over 75% of patients with acute myocardial
infarctions have abnormal LVEFs. Patients showing
a serial decline in LVEF have a significantly higher risk of
mortality in the early postinfarction period. Inferior
myocardial infarctions have associated right ventricular
wall motion abnormalities in up to 40% of patients.
In current practice, SPECT myocardial perfusion
scintigraphy is preferred over RVG for evaluation of coro-
nary artery disease.
Figure 14-54 Exercise induced global hypokinesis. Note the
decreased emptying of the left ventricle in response to exercise
(arrows) compared to the rest study. ED, End diastolic; ES, end
systolic; EX, exercise.This is characteristic of apical ischemia.
502 NUCLEAR MEDICINE: THE REQUISITES
Box 14-23 Cause of Abnormal
Ventricular Functional
Response to Exercise
Hemodynamically significant coronary artery disease
Cardiomyopathy
Myocarditis
Valvular heart disease
Pericardial disease
Drug toxicity
Prior surgery or injury
Valvular Heart Disease
Patients with valvular heart disease can have pressure
overload, volume overload, or both. The response to
pressure overload from stenosis is concentric hypertro-
phy. The response to volume overload from insuffi-
ciency is dilatation. Pulmonary vascular hypertension
and eventually congestive heart failure are the result of
significant valve disease. RVG allows assessment of ven-
tricular size and ejection fraction. Because the ejection
fraction is in part determined by preload, afterload, and
heart rate, resting ejection fraction cannot be used alone
to assess myocardial contractility or functional reserve.
Patients with valvular heart disease may have regional or
global dysfunction that cannot be differentiated from
coronary artery disease.
The findings regarding chamber size and function on
RVG are similar to cardiac catheterization. A measure-
ment that is easier to determine with RVG than with
contrast angiography is calculation of stroke volume
ratios for the left and right ventricles. With mitral insuf-
ficiency, for example, some of the blood is propelled
antegrade and some regurgitates through the mitral
valve during each left ventricular contraction. In nor-
mal subjects, the stroke–volume ratio between the ven-
tricles should be 1.0 because all of the blood is
propelled antegrade. Stroke–volume ratio provides
a measure of the severity of regurgitation that can be fol-
lowed sequentially.
The major limitation of the calculation of the stroke
volume ratio from equilibrium blood pool studies is
chamber overlap between the right and left ventricles
and between the right ventricle and the right atrium.
The exact level of the pulmonic valve is also difficult to
establish in many cases. For these reasons a LV/RV ratio
of 1.5 is used as the upper limit of normal, which is
greater than the expected value of 1.0.
Cardiomyopathy and Myocarditis
Cardiomyopathies may be classified as congestive, hyper-
trophic, or restrictive. In congestive cardiomyopathies,
the ventricles are typically enlarged and dysfunctional.
The global ejection fraction is decreased and wall motion
is uniformly poor, except that the septal and anterior
basal segments are frequently spared. The hallmark of
the hypertrophic cardiomyopathies is asymmetrical
septal hypertrophy. Echocardiography is the diagnostic
procedure of choice. The left ventricular chamber is typ-
ically small, and the LVEF is above normal. Diastolic fill-
ing is abnormal because of poor compliance of the
hypertrophied myocardium.
Assessment of Cardiac Toxicity
The most common indication for RVG in current prac-
tice is to assess the potential cardiotoxic effects of non-
cardiac drugs. Anthracycline drugs used most
commonly in the treatment of breast cancer and malig-
nant lymphoma, such as doxorubicin (Adriamycin), pro-
duce a cumulative dose-dependent depression of left
ventricular function.
Heart failure develops in 4–20% of patients who receive
a cumulative dose of >500 mg/ m2, in 18% of those receiv-
ing >550 mg/m2, and in 36% receiving >600 mg/m2.
Acute hemodynamic decompensation may be followed
by an irreversible dilated cardiomyopathy with LV dys-
function. Endocardial biopsy is diagnostic but invasive
and not commonly performed.
Overt congestive heart failure is preceded by a progres-
sive fall in the LVEF. Serial monitoring of left ventricular
function can detect a change in cardiac function over time
and the drug can be stopped or reduced when a reduction
in LVEF is observed. The incidence of overt cardiac failure
can be significantly reduced. Complete recovery may
occur if therapy is discontinued at an early stage.
Both the absolute LVEF and the magnitude of fall are
important. Generally, a 10% decline in LVEF to below the
lower limit of normal (50%), an absolute LVEF of 40% or a
20% decline in LVEF at any level is indicative of deteriora-
tion in cardiac function.
The resting RVG is insensitive for early detection
compared to endocardial biopsy. Limited data suggests
an exercise RVG study in addition to rest sensitivity for
detection. However, the specificity is low without
serial testing. Furthermore, many of these patients are
debilitated and cannot exercise adequately. Resting
studies are the routine in most clinics. Noncardiac con-
ditions can also affect the LVEF, including anemia, fever,
and sepsis.
Pulmonary Disease
Right ventricular enlargement can be diagnosed with RVG.
In patients with a new onset of dyspnea, the RVG can help
differentiate left ventricular from pulmonary dysfunction.
The demonstration of a normal left ventricular ejection
fraction, wall motion, and chamber size strongly suggests
a pulmonary etiology.
 Cardiac System 503

central circulation is studied using the first-transit tech-

nique. Early recirculation into the right ventricle as
blood flow bypasses the lung is detected with a curve-
fitting technique. In brief, the lung transit curve ( Fig.
14-55A) is modeled by a mathematical function called a
gamma variate ( Fig. 14-55B). The contribution to the
time-activity curve from recirculation is taken as the dif-
ference between the total area under the time-activity
curve minus the area under the gamma variate fit
( Fig. 14-55C). This approach allows detection of shunts
as small as 20%.
Right-to-left shunts may be detected and quantified
0 10 20 30 40 with Tc-99m-labeled macroaggregated albumin. The
ratio of tracer in the lung to tracer gaining access to the
A Time (sec)
systemic circulation provides a measure of the severity
of shunting. With right-to-left shunts, images show
uptake with the cerebral cortex and other organs.
Quantification is best done with whole-body imaging.
Regions of interest can be drawn for the lungs and total
body. The calculated percent shunt does not accurately
reflect the real percent shunt. Greater than 10% is
abnormal. Brain uptake is always seen.
Right-to-left shunts are generally given as a relative
1 2 contraindication to the use of macroaggregated albumin,
owing to the theoretical risk of embolizing the capillary
bed of the brain. In practice, this event has not been
a problem. However, it is recommended that the num-
0 10 20 30 40
ber of particles be reduced.
B Time (sec)

INFARCT-AVID IMAGING

Technetium-99m Pyrophosphate
Radiolabeling Tc-99m pyrophosphate ( Tc-99m PYP) was
originally used as a bone scan radiopharmaceutical and is
prepared in the same manner. Sodium pertechnetate
from a generator is added to a vial containing stannous
pyrophosphate Sn ( II), a reducing agent. The Tc-99m
forms a chelate with the pyrophosphate molecule.

0 10 20 30 40
C Time (sec)
Figure 14-55 Left-to-right shunt calculation A, Time–activity
curve obtained from a region of interest over the lungs in a patient
with a left-to-right shunt.The second peak is due to early
recirculation of tracer through the left-to-right shunt. B, The
relative contributions from the initial transit and the shunt are
determined from a curve-fitting technique. C, Initial time–activity
curve and the two mathematically fitted curves.The shunt ratio
(Qp/Qs) is calculated from the areas under these curves.
Congenital Heart Disease
Right-to-left shunts may occur due to a variety of con-
genital cardiac diseases. For left-to-right shunts, the
Mechanism of Localization
After cell death in acute myocardial infarction, an influx of
calcium occurs and calcium phosphate complexes are
formed. These microcrystalline deposits act as sites for
Tc-99m PYP uptake. Some binding may also occur
on denatured macromolecules. The status of the peri-
infarction circulation is important in tracer uptake. Some
residual blood flow is necessary to deliver the tracer to the
infarct area and surrounding tissue. The tracer then dif-
fuses into the necrotic tissue and is bound. Highest
uptake is at the periphery of infarction. In large infarc-
tions, with neither direct flow nor diffusion to the central
area, no tracer is delivered and a characteristic ring or
doughnut pattern is seen due to activity around the mar-
gin of the damaged area.
504 NUCLEAR MEDICINE: THE REQUISITES
Methodology
Imaging is performed 3 to 4 hours after tracer administra-
tion to allow clearance from blood pool. Radioactivity
retained in the circulation contributes to background
blood pool activity, which can be confused with myocar-
dial uptake; false-positive interpretations of Tc-99m PYP
images could result. A further delay to allow more com-
plete clearance should be considered if background
activity remains high 4 hours after injection.
A standard protocol is described ( Box 14-24).Tc-99m
pyrophosphate is injected intravenously, 555–925 MBq
( 15–25 mCi). A high-resolution collimator should be
used. Multiple planar views and/or SPECT are per-
formed. SPECT offers greater image contrast, allowing
detection of smaller abnormalities and more exact
anatomical localization of infarct. In general, the tech-
nique is used only when other clinical parameters are
nondiagnostic.
The sensitivity for detecting infarction is highest at
24–48 hours after the acute event. A major limitation
for diagnosis of acute myocardial infarction is the delay
between the time of infarction and the time of scinti-
gram positivity. Significant uptake becomes demon-
strable at 12 hours after infarction. Maximum loca-
lization occurs at 48 to 72 hours. Thereafter uptake
begins to diminish as the infarcted area heals. In
uncomplicated cases, the scintigram reverts to normal
within 14 days. If initial images reveal diffuse activity
in the region of the heart, further delay can be helpful
to allow more complete clearance of tracer from the
blood pool.
Scintigraphic Patterns in Acute Myocardial
Infarction
Tc-99m PYP is an avid bone seeker. In normal subjects
and in patients without infarction, the sternum and ribs
are clearly seen, with no focal or diffuse activity in the
region of the heart. Faint cardiac blood pool may be
seen.
The classic scintigraphic pattern in myocardial infarc-
tion is a focal area of increased tracer uptake correspond-
ing to the affected region of the heart. A grading system
is sometimes used: zero for a normal study, 1+ for faint
uptake, 2+ for uptake equal to rib intensity, and 3+
for uptake greater than rib intensity. Diagnostic confi-
dence increases with the grade of uptake and with focal
versus diffuse activity.
Complete interpretation includes assessment of loca-
tion and size. Location is inferred from comparison of
the relationship of the abnormal uptake to the expected
location of the heart and the skeletal structures on multi-
ple planar views or SPECT. Anterior infarctions are seen
en face on the anterior view and project just behind the
sternum on the lateral view ( Fig. 14-56). Lateral wall
infarcts appear as vertical curvilinear lesions on the ante-
rior view. With progressive obliquity, the area of abnor-
mality moves either closer to the sternum (anterolateral
infarcts) or farther from the sternum (posterolateral
infarcts). Inferior wall infarctions are concave upward
and may have a characteristic “lazy 3” configuration if
they involve the inferior septum and right ventricle.
Large infarctions, most frequently in the anterior wall
of the left ventricle, may exhibit a doughnut pattern of
increased uptake resulting from absence of tracer in the
center of the infarct area. This pattern is associated with
a poor clinical prognosis. A minimum of 3 g of tissue
must be infarcted for scintigraphic detection.
Sensitivity is high for transmural or Q-wave infarc-
tions, approximately 95%. For subendocardial infarc-
tions,the sensitivity is considerably less,probably 65% for
planar scintigraphy. Specificity is greater than 90%.
Normal rib uptake may obscure small regions of uptake.
SPECT can improve sensitivity and localization.
There are numerous causes of false positive Tc-99m
PYP scans ( Box 14-25), including diffuse activity in the
cardiac blood pool misinterpreted as emanating from the
myocardium, uptake in areas of chest wall trauma, in
skeletal muscle that is necrotic because of prior car-
dioversion, and in calcifications in or near the heart.
Calcifications in the costal cartilage may have uptake.
Several conditions result in diffusely increased
myocardial uptake of Tc-99m pyrophosphate. The most
dramatic is amyloidosis ( Fig. 14-57). The tip-off to amy-
loid as the etiology is visualization of the entire
myocardium, including the right ventricle, with quite
Box 14-24 Protocol: Tc-99m
Pyrophosphate Infarct-Avid
Imaging
PATIENT PREPARATION AND FOLLOW-UP
EKG leads should be moved out of field of view
Frequent voiding to minimize radiation dose to bladder
DOSAGE AND ROUTE OF ADMINISTRATION
740 MBq (20 mCi) Tc-99m pyrophosphate intravenously
TIME OF IMAGING
3–4 hr after radiopharmaceutical administration (may be
performed at 1 hr if clinically indicated)
PROCEDURE
Collimator: low-energy, high-resolution or general
purpose, parallel hole collimator
Planar images:Anterior view for 500k counts and record
the length of time, 35˚ left anterior oblique (LAO), 70˚
LAO, and left lateral views for equal time
SPECT
 Cardiac System 505

Figure 14-56 Lateral wall myocardial infarction with Tc-99m pyrophosphate uptake.The amount
of myocardial uptake is greater than rib uptake and not equal to sternal uptake.

good myocardium-to-background ratio. Myocarditis,
postradiation injury, and doxorubicin cardiotoxicity are
all reported causes of diffusely increased myocardial
uptake.
Tc-99m PYP scintigrams may remain abnormal for
weeks or months after a myocardial infarction. Those
scintigrams that continue to show uptake for more than
3 months correlate with a higher risk for future infarction.
The Fab′ fragment is radiolabeled with In-111 or Tc-99m.
The sensitivity for detecting acute myocardial infarc-
tion is high, over 85%. A major advantage is that there
is no rib uptake. A disadvantage is the slow pharmaco-
kinetics of antimyosin antibody, which means that opti-
mum imaging cannot be accomplished for many hours
after radiopharmaceutical administration because of
high background activity. Uptake may also be seen
with myocarditis, cardiac transplant rejection, and drug
toxicity.
Clinical Applications and Utility
The major limitation of infarct-avid scintigraphy is its
delayed positivity after the onset of symptoms. In most
patients, the diagnosis is established from the history,
physical examination, ECG, and serum enzyme determi-
nations before the optimal time window for imaging.
The study is used mainly when the diagnosis is uncertain
( Box 14-26).
Tc-99m pyrophosphate and In-111 antimyosin anti-
body imaging has been used in patients with conduction
system abnormalities on the resting ECG (e.g.,left bundle
branch block) that limited interpretation, and in patients
506 NUCLEAR MEDICINE: THE REQUISITES
Box 14-25 False-Positive Tc-99m
Pyrophosphate Infarct-Avid
Studies
FOCAL
Old myocardial infarction (persistent positivity)
Calcification: valvular, pericardial
Ventricular aneurysm
Costal cartilage calcification
DIFFUSE
Myocarditis
Pericarditis
Cardiomyopathy
Amyloidosis
Radiation therapy
Persistent blood pool activity
Doxorubicin (Adriamycin) therapy
who presented late after symptom onset when cardiac
enzymes were not definitive of availability of better
serum markers and imaging techniques; hot spot infarct
imaging is used infrequently today. Tc-99m PYP imaging
still may have a role in the diagnosis of amyloidosis.
INVESTIGATIONAL CARDIAC
RADIOPHARMACEUTICALS
I-123 Meta-Iodobenzyl-Guanidine
I-123 MIBG has been used to study the adrenergic status
of the heart. The heart is richly innervated and MIBG
has been used to provide some interesting insights.
Uptake of MIBG is blocked in patients taking drugs, such
as guanethidine and cocaine, that compete for uptake
into the presynaptic storage vesicles of the adrenergic
system. Decreased uptake is seen after myocardial
Figure 14-57 Cardiac amyloidosis.Tc-99m pyrophosphate is
taken up throughout the left ventricular myocardium. Subtle
uptake can also be seen in the right ventricular myocardium.
Box 14-26 Indications for Tc-99m
Pyrophosphate Scintigraphy
Suspected infarction in a patient with left bundle-branch
block
Delay in diagnosis; enzymes past expected peak
After cardioversion
After major surgery or trauma
Subendocardial (non-Q-wave) infarction versus ischemia
Baseline ECG abnormal because of prior myocardial
infarctions
Prior ECG not available
Right ventricular infarction
infarction and in diabetic patients with denervated
hearts. Some patients with cardiomyopathies also have
diminished or absent uptake. A clinical role has not
been established for MIBG, although it is being used to
assess reinnervation after cardiac transplantation and
to help determine prognosis in patients with dilated
cardiomyopathy.
Fatty Acid Radiopharmaceuticals
Fatty acids supply the majority of the heart’s metabolic
requirements under normal aerobic conditions. With
ischemia, energy metabolism shifts to anaerobic metabo-
lism and the main energy substrate changes from free fatty
acids to glucose metabolism. Radiolabeled fatty acids can
be used to image myocardial aerobic metabolism.
PET imaging with carbon-11 palmitic acid gives
information similar to perfusion agents. Myocardial
time-activity curves reflect fatty acid metabolism.
Decreased uptake and delayed clearance indicates
ischemia. C-11 acetate also provides good images of
the heart. Because of its rapid metabolism, cardiac
turnover is inferred from dynamic analysis of its
myocardial clearance pharmacokinetics. Controversy
exists regarding the significance of the metabolic infor-
mation provided. Neither agent is FDA-approved and
an onsite cyclotron is required.
I-123 BMIPP (β-Methyl-p-iodophenylpentadecanoic
acid) is a newly investigated single-photon branching
free fatty acid radiopharmaceutical with slow metabo-
lism; thus, it is well-suited for SPECT. With ischemia,
there is decreased uptake similar to Tl-201. The unique-
ness of this radiopharmaceutical is that it can demon-
strate a persistent disturbance of fatty acid metabolism,
even when blood flow has been reestablished, such
as with unstable angina and stunned myocardium.
Similar metabolic abnormalities have been described in
cardiomyopathies.

SUGGESTED READING
Books
Botvinick EH: Nuclear Medicine Self-Study Program III: Nuclear
Medicine Cardiology,Topic 1-6. Society of Nuclear Medicine,
1997.
DePuey EG, Garcia EV, Berman DS: Cardiac SPECT Imaging, 2nd
ed. Philadelphia, Lippincott Williams & Wilkins, 2001.
Dilsizian V, Narula J:Atlas of Nuclear Cardiology. Philadelphia,
Current Medicine, 2003.
Gerson MC: Cardiac Nuclear Medicine, 3rd ed. New York,
McGraw-Hill, 1997.
Zaret BL, Beller GA: Clinical Nuclear Cardiology, 3rd ed.
Philadelphia, Elsevier, 2005.
Reviews
Bacharach SL, Bax JJ, Case J, et al: PET myocardial glucose metabo-
lism and perfusion imaging: Part 1—Guidelines for patient prepa-
ration and data acquisition. J Nucl Cardiol 10:543–554,2003.
Baird MG, Bateman TM, Berman DS: Guidelines for the clinical
use of cardiac radionuclide imaging. J Am Coll Cardiol 2003.
Cerqueira MD, Weissman NJ, Disizian V, et al: Standardized
myocardial segmentation and nomenclature for tomographic
imaging of the heart. Circulation 105:539–549, 2002.
Freeman LM, Blaufox MD: Cardiovascular nuclear medicine,
parts 1 and 2. Semin Nucl Med 19, 1999.
Hendel RC, Corbett JR, Cullom SJ, et al: The value and practice
of attenuation correction for myocardial perfusion SPECT
Imaging: a joint position statement from the American Society
of Nuclear Cardiology and the Society of Nuclear Medicine.
J Nucl Med 43:273–280, 2002.
Cardiac System 507
Mieres JH, Shaw LJ, Hendel RC, et al:A report of the American
Society of Nuclear Cardiology task force on women and heart
disease. J Nucl Card 10:1–11, 2003.
Shelbert HR, Beanlands RB, Engel F, et al: PET myocardial perfu-
sion and glucose metabolism imaging. Part 2—Guidelines for
interpretation and reporting. J Nucl Card 10: 557–571, 2000.
Publications
Arrighi JA, Soufer R: Reverse redistribution: is it clinically rele-
vant or a washout? J Nucl Cardiol 5:195–201, 1998.
Bax JJ, Cornel JH, Visser FC, et al: Comparison of fluorine-
18-FDG with rest-redistribution thallium-201 SPECT to delin-
eate viable myocardium and predict functional recovery after
revascularization, J Nucl Med 39:1481–1486, 1998.
Braunwald E, Rutherford JD: Reversible ischemic left ventricu-
lar dysfunction: evidence for the “hibernating myocardium,”
J Am Coll Cardiol 8:1467, 1986.
DePuey G, Parmett S, Ghensi M, et al: Comparison of Tc-99m ses-
tamibi and Tl-201 gated SPECT, J Nucl Cardiol 6:278–285, 1999.
Hansen CL, Rastogi A, Sangrigoli R: On myocardial perfusion,
metabolism and viability, J Nucl Cardiol 5:202–205, 1998.
Manrique A, Foraggi M, Vera P, et al: Tl-201 and Tc-99m MIBI
gated SPECT in patients with large perfusion defects and left
ventricular dysfunction:compression with equilibrium radionu-
clide angiography, J Nucl Med 40: 805–809, 1999.
Merlet P, Pouillart F, Dubois-Rande J, et al: Sympathetic nerve
alterations assessed with I-123-MIBG in the failing human heart,
J Nucl Med 40:224–231, 1999.
Santana-Boado C, Candell-Riera J, Castell-Conesa J, et al:
Diagnostic accuracy of technetium-99m-MIBI myocardinal
SPECT in women and men, J Nucl Med 39:751–755, 1988.
Figure 14-A 1, Inferior wall ischemia. Exercise Tc-99m sestamibi/rest Tl-201 myocardial
perfusion. Marked hypoperfusion of the entire inferior wall and inferior apex post-stress, which
normalizes on rest images consistent with a large region of severe ischemia.
Figure 14-A 2, Inferior wall ischemia: polar map and volume quantitative display for A1.
The reversibility perfusion (%) box shows the extent and severity of the reversible perfusion
defect as a polar map and three-dimensional volume display. At the right, the stress extent (%)
and reversibility (%) are shown in graph form.
Figure 14-B 1, Anterior wall ischemia. Postexercise and 3-hour delayed Tl-201 myocardial
perfusion. Stress induced hypoperfusion is seen in the anterior wall extending to the apex.At rest,
there has been near complete redistribution consistent with moderate ischemia of a moderate sized
region.This subjective evaluation can be compared to the quantitative display in B2.
Figure 14-B 2, Anterior wall ischemia: polar map and volume quantitative display. Reversibility
of 10–13% in individual regions of the anterior lateral apical region.The graph in the right hand
column displays the stress extent and reversibility percentage.This suggests a somewhat lesser
degree of ischemia than the interpretation of the splash display in B1.
Figure 14-C Attenuation correction: attenuation artifact. Persantine stress and rest Tc-99m
sestamibi.Top two rows (stress, rest) of transverse, vertical and horizontal long axis slices are non
attenuation corrected (NAC) images. Lower two rows of each section are attenuation corrected
(AC).The NAC images show decreased activity in the inferior wall.With AC, the distribution of
perfusion is normal.This is a normal stress myocardial perfusion study.The patient had a history of
arrhythmia and some risk factors for coronary disease, but no history of angina or myocardial
infarction.
Figure 14-D Attenuation correction: inferior wall infarction. Persantine stress and rest Tc-99m
sestamibi.Top two rows of transverse, vertical and horizontal long axis slices are non attenuation
corrected images. Lower two rows of each section are attenuation corrected.The stress (above) and
rest (below) non-attenuation corrected images show a large area of decreased activity in the inferior
wall extending to the septum and lateral wall.Attenuation correction images show persistent
decreased activity in the same region, although to a lesser degree.The study is consistent with large
area of moderately severe infarction in the inferior wall.This patient had a history of prior
myocardial infarction.
Figure 14-E 1, Multivessel ischemia. Exercise stress Tl-201 and reinjection rest Tl-201.The
patient had known three-vessel coronary artery disease and poor ventricular function, and was
beginning considered for coronary bypass surgery.With submaximal stress, there is reduced
perfusion to the anterior, septal, and inferior walls. On delayed imaging, there is redistribution to
these regions.There is a suggestion of transient ischemic dilation (TID). Stress-induced lung uptake
was also seen (see Fig. 14-19).
Figure 14-E 2, Multivessel ischemia: polar display. The quantitative two- and three-dimensional
displays show redistribution primarily to the apex, inferior-septal and septal regions in the range of
11–17%.This display likely underestimates the degree of ischemia, when interpreted in light of
submaximal exercise,TID, and exercise induced lung uptake (see Fig. 14-19).
Figure 14-F 1, Anterior lateral and inferior lateral wall ischemia. Persantine stress and rest Tl-201.
The stress images show hypoperfusion of the anterior lateral wall and no perfusion to the inferior
wall. Rest images show definite redistribution to the anterior lateral wall but incomplete
redistribution to the inferior wall.
Figure 14-F 2, Lateral wall ischemia: polar and volume quantitative display. Reversibility
percentage is as high as 23% to 32% in the anterior lateral and inferior lateral walls. See wall motion
and thickening on Fig. 14-K.
Figure 14-F 3, Lateral wall ischemia: wall motion and thickening.The images in the left column
show septal thickening (manifested as brightening) and to a lesser degree anterior lateral wall
thickening.There is reduced thickening in the inferior wall.The cardiac volume curve and calculated
ejection fraction (41%) show diffuse hypokinesis.
Figure 14-G 1, Lateral wall infarction and inferior lateral ischemia. Persantine stress rest Tl-201.
Stress images show no perfusion of the lateral wall and reduced perfusion to the inferior wall. Rest
images show no improvement in the lateral wall but improved perfusion to the inferior wall.
Figure 14-G 2, Lateral wall infarction and inferior lateral ischemia: polar and volume
quantitative display.The reversibility percentage shows at least one inferior wall region with a 19%
reversibility.The graph in the right column shows the discrepancy between the stress extent and the
reversibility extent, representing the infarcted lateral wall.
 15
CHAPTER Pulmonary System

Pulmonary Embolism
Ventilation Perfusion Scintigraphy PULMONARY EMBOLISM
Radiopharmaceuticals
Perfusion Radiopharmaceuticals Pulmonary thromboembolism is a common clinical
Ventilation Radiopharmaceuticals problem. Although the true incidence of clinically sig-
Radioactive Gases nificant emboli is difficult to assess, pulmonary emboli
Radioaerosols have been found in up to 70% of autopsies. Some
Dosimetry authorities have estimated an annual incidence of
Technique 650,000 cases per year with over 100,000 deaths.
Xenon-133 Ventilation Untreated pulmonary embolism ( PE) is frequently fatal.
Tc-99m DTPA The mortality rate of approximately 30% can be
Tc-99m MAA reduced to 3–10% with anticoagulation therapy or infe-
Image Interpretation rior vena cava filter placement. Predisposing factors
Perfusion Scintigram include immobilization, recent surgery, underlying
Ventilation Scintigram malignancy, and various hypercoagulable states. In
Differential Diagnosis of Ventilation Abnormalities women, pregnancy and estrogen use are considered
Image Interpretation for Pulmonary Embolism Diagnosis risk factors.
PIOPED Criteria The clinical diagnosis of pulmonary embolism is diffi-
Normal cult as presenting symptoms are nonspecific. The clas-
High Probability sic triad of dyspnea, pleuritic chest pain, and hemoptysis
Low/Very-Low Probability is rarely encountered. Patients often complain of short-
Intermediate Probability ness of breath, chest pain, and cough. Tachycardia is fre-
Special Signs quently present and, occasionally, patients present with
Approach to Interpreting the Ventilation-Perfusion Scintigram cor pulmonale and circulatory collapse.
Accuracy of Ventilation-Perfusion Scintigraphy Arterial blood gases often show evidence of respira-
Differential Diagnosis of V/Q Mismatches tory alkalosis and a low PAO2. However, the PAO2 may be
Computed Tomography Pulmonary Arteriography (CTPA) normal in some patients. The plasma D-dimer is an excel-
for the Diagnosis of Pulmonary Emboli lent screening test with a low false-negative rate.
Other Applications of Ventilation-Perfusion Scintigraphy However, it is nonspecific and can be elevated in various
Quantitative Lung Scan inflammatory conditions as well as in PE.
Adult Respiratory Distress Syndrome Pulmonary arteriography has been considered the
Detection of Deep Venous Thrombosis “gold standard” for PE diagnosis with a reported sensitiv-
ity of 98% and specificity of 97%. The arteriogram offers
excellent imaging resolution as well as the ability to per-
form direct vascular pressure measurements. However,
readings vary among interpreters, particularly for distal
emboli. Although the risk from the procedure is rela-

508

tively low, it is an invasive test with a risk of death of
approximately 0.5%. Therefore, less invasive imaging
tests are more widely used.
Because there is a definite association between deep
vein thrombosis ( DVT) and PE, ultrasound examination
of the leg veins is often indicated. Sonography has
a reported sensitivity of 94% and specificity of 99% for
DVT in the thigh veins but is much less sensitive for eval-
uating pelvic veins or the veins below the knee. The
presence or absence of DVT does not determine
whether or not PE is present.
Several radiographic signs of PE on chest x-ray have
been described; however, these signs are rarely present.
Westermark’s sign is diminished vascularity in an area
affected by a PE. The Fleischner sign is dilation of the
pulmonary artery proximal to the embolus. Signs of
infarction following a pulmonary embolus include an
infiltrate from hemorrhage filling the alveolar spaces and
the pleural-based Hampton’s hump. The most common
x-ray findings in pulmonary embolus are not specific for
PE. Pleural effusions are common, present as often as
50% of the time. Infiltrates and atelectasis are also com-
mon. Normal radiographs have been described in any-
where from 12–30% of patients with PE. Although
a chest radiograph is not adequate for PE diagnosis, an
x-ray should always be obtained as it can identify problems
that might present with similar symptoms to an embolus.
Computed tomography pulmonary angiography
(CTPA) is now widely used as a method for PE diagnosis.
In recent years, developments in spiral computed tomog-
raphy (CT) have enabled rapid imaging techniques. The
central pulmonary arteries can be visualized during con-
trast opacification with a fairly high degree of reliability.
Acute clots and changes from chronic PE can be visual-
ized. At the same time, CT can identify other problems
and nonembolic causes for the patient’s symptoms.
Technical problems that make CT nondiagnostic include
motion artifacts and insufficient contrast opacification.
Contrast CT is often contraindicated in patients with iodin-
ated contrast allergies or renal insufficiency.
Magnetic resonance ( MR) offers the ability to image
vascular structures without ionizing radiation or iodin-
ated contrast. Currently,gradient-echo and spin-echo MR
techniques are used for several vascular and cardiac
applications. More powerful gradients and faster imag-
ing sequences now allow imaging of the entire chest dur-
ing a single breath-hold, and new contrast-enhanced
techniques have yielded excellent pulmonary arteriogra-
phy images. Preliminary results are quite good, although
technical problems such as cardiac motion artifact
obscuring lower lobe arteries persist. MR may play
a clinical role in PE diagnosis in the future.
The ventilation perfusion ( V/Q) scintigram remains
an important diagnostic tool for PE. It is noninvasive,
Pulmonary System 509
low risk, and time proven. V/Q scans are often used for
patients where CT contrast is contraindicated or when
patient radiation exposure is a concern, such as preg-
nancy. A typical CTPA protocol results in an exposure
on the order of 8–10 mSv, but a V/Q scan has a fraction
of the exposure at about 2 mSv.
VENTILATION PERFUSION
SCINTIGRAPHY
Ventilation (V) and perfusion (Q) in the lungs are normally
coupled or matched. A normal functional gradient is seen,
with the lung apices receiving less perfusion and ventila-
tion than the lung bases. Areas of the lung which are well
ventilated are also normally well perfused (Fig.15-1).
Many disease processes affect aeration. Among these
are obstructive pathologies such as asthma, bronchitis,
bronchiectasis, and emphysema. Acutely, the normal
regional response to ventilatory causes of hypoxia is
vasoconstriction, which shunts blood flow away to
other aerated regions. This results in regions of reduced
but matched ventilation-perfusion abnormalities and is
often seen with asthma. Scarring results in matched
abnormalities in chronic disease such as emphysema.
Often, the chest radiograph will reveal a reason for
abnormal aeration and perfusion such as an infiltrate or
mass. This is often described as a “triple match”
with abnormal ventilation, perfusion, and radiographic
findings.
A vascular abnormality, such as pulmonary embolus,
reduces pulmonary arterial perfusion. Commonly, lung
parenchyma remains viable in these cases due to the
bronchial arterial system. Therefore, the normal alveolar
spaces will remain aerated, thus preventing infarction.
Ventilation and perfusion are uncoupled or mismatched
in the regions affected by the pulmonary embolus.
These physiologic parameters form the foundation
for the ventilation perfusion lung scintigram. The perfu-
sion portion of the scan uses intravenously injected
microscopic particles large enough to be trapped on the
first pass through the pulmonary capillary bed. When
a radioactive label is attached to the particle, the distri-
bution of microemboli can be imaged and present a pic-
ture of blood flow distribution within the lungs. To
determine if a perfusion defect is the result of a ventila-
tion abnormality rather than a primarily vascular prob-
lem, a ventilation study is done with an inhaled
radiopharmaceutical ( Box 15-1). By imaging pulmonary
perfusion and lung ventilation, areas of matched and
mismatched defects can be identified. As most mis-
matched perfusion defects are the result of pulmonary
emboli, a diagnosis can be made with a high degree of
probability.
510 NUCLEAR MEDICINE: THE REQUISITES

Figure 15-1 Normal V/Q scan. Ventilation (A) and perfusion (B) lung scans show homogeneous
radiotracer distribution and the normal gradient of increasing activity in the bases relative to the
apices.

the first pass, the particles must be larger than capillary

Radiopharmaceuticals
Perfusion Radiopharmaceuticals
The diameter of a red blood cell is just under 8 μm. The
capillaries are slightly larger, ranging from 7–10 μm and
precapillary arterioles are on the order of 35 μm. For
a perfusion agent to be trapped in the capillary bed on
size. However, if the particle is too large, distribution
may not fully reflect perfusion as particles become
trapped in the more central arterioles rather than capil-
laries and precapillary arterioles.
Several different particulate agents have been used
over the years. The first studies in humans were

Box 15-1 Ventilation and Perfusion
Agents of Historical Interest
Perfusion Ventilation
Tc-99m human albumin RADIOACTIVE GASES
microspheres Xenon-133*
Tc-99m macroaggregated Xenon-127
albumin* Krypton-81m
RADIOAEROSOLS
Tc-99m DTPA*
Tc-99m Technegas
*Currently commercially available in the United States.
obtained with radioiodinated ( I-131) macroaggregated
albumin. Radiolabeled technetium-99m agents were
subsequently introduced into clinical use. One of these,
Tc-99m-labeled human albumin microspheres ( Tc-99m
HAM) provided rigid nondegradable particles of uniform
size. However, it is no longer available in the U.S. for
human use.
Tc-99m macroaggregated albumin ( Tc-99m MAA) is
the radiopharmaceutical used today. The preparation
contains particles ranging in size from 5–100 μm. The
majority ( 60–80%) are in the 10–30 μm range. A kit is
available that contains MAA with stannous ion. Tc-99m
as sodium pertechnetate is added to the reaction vial,
resulting in rapid labeling of the MAA particles.
After intravenous injection into a peripheral vein,
the radiolabeled particles travel through the right atrium
and right ventricle where mixing occurs. They are then
filtered out or trapped as they travel through the pulmonary
bed. In areas of decreased or absent perfusion,
correspondingly fewer particles are delivered and trapped,
resulting in relatively photopenic or “cold” areas. This
produces a scintigraphic map of relative perfusion in
the lungs.
The MAA particles have a biological half-life in the
lungs of approximately 4–6 hours. Over time, the par-
ticles undergo mechanical degradation and fragment.
They may then lodge in smaller vessels but eventually
gain access to the circulation where they are phagocy-
tosed in the reticuloendothelial system. The number
of particles used is an important consideration. If the
dose of particles is too small, the distribution pattern in
the capillary bed will not be statistically valid. A mini-
mum of 100,000 particles is required in adults. On the
other hand, injecting too many particles could
theoretically obstruct a hemodynamically significant
portion of the pulmonary circulation. In practice,
Pulmonary System 511
200,000–500,000 particles are administered. Because
there are an estimated 300 million precapillary arteri-
oles and over 280 billion pulmonary capillaries, this
should result in obstruction of only a small fraction of
vessels, between 0.1–0.3%. Most of these are only par-
tially blocked.
The perfusion exam is generally very safe. Only a tiny
fraction of capillaries are occluded and most of these are
only partially blocked. However, certain conditions war-
rant reducing the number of particles. These include
pulmonary hypertension, pregnancy, and right-to-left car-
diac shunts. The number of particles is also reduced for
neonates ( 10,000) and children under the age of 5 years
(50,000–150,000).
Patients with pulmonary hypertension may have sig-
nificantly fewer capillaries than normal. Theoretically,
their condition could worsen if too many were
occluded. Although significant effects are rare, the num-
ber of particles administered is often reduced to
100,000–250,000.
Patients with right-to-left cardiac shunts also require
caution. The Tc-99m MAA will pass through the heart
into the systemic circulation rather than remain in the
lungs. The kidneys, brain, and other structures will be
visualized. Although the idea of microemboli lodging in
the brain is alarming, it does not seem to have clinical sig-
nificance. Tc-99m MAA has been long used to diagnose
right-to-left cardiac shunts and to determine their severity.
This is done by calculating the amount of activity trapped
in the lungs versus the amount of activity that bypasses
the lungs. Investigators have also injected particles into
other organs, including the brain, to study blood flow
without ill-effect. Although it is prudent to decrease the
number of particles to 100,000–150,000, a known right-
to-left shunt is not an absolute contraindication.
When PE is a clinical consideration in pregnant
patients, a V/Q exam offers the ability to diagnose the
emboli with relatively low radiation to the fetus.
However, the minimum radiation dose should be used.
This can be accomplished by decreasing the amount of
Tc-99m MAA used, but it is essential that the dose con-
tains the minimum 100,000 particles.
When ordering the radiopharmaceutical, it must be
made clear that a lower number or particles is desired
and not a lower level of radioactivity. To achieve the
right balance between the amount of radioactivity and
the number of particles, the amount of Tc-99m pertech-
netate added to the reaction vial may need to be
adjusted. Because commercially available reaction
vials provide for multiple doses, the radioactivity-to-
particles ratio changes continuously after the initial
labeling. This should be borne in mind in making dose
calculations.
512 NUCLEAR MEDICINE: THE REQUISITES

Ventilation Radiopharmaceuticals
Two classes of radiopharmaceuticals are available for Table 15-1 Comparison of Xenon-133 and Tc-99m
ventilation imaging: radioactive gases and radioaerosols DTPA Ventilation Agents
(Box 15-1). One aerosol,Tc-99m diethylenetriamine pen-
taacetic acid ( Tc-99m DTPA), and one gas, xenon-133 Xenon-133 Tc-99m DTPA
( Xe-133), are currently available in the U.S. Comparative
Mode of decay Beta-minus Isomeric
studies between Xe-133 and Tc-99m DTPA have not Physical half-life 5.3 days 6 hours
shown significant differences in overall accuracy. The Biological half-life 30 seconds 45 minutes
choice of agent depends largely on available equipment, Photon energy 81 keV 140 keV
proper room ventilation, the frequency of chronic Multiple-view imaging No Yes
obstructive pulmonary disease ( COPD) in the referral Useful for severe COPD Yes +/−
Used after perfusion scan No No
population, and personal preference.

Radioactive Gases After lung inhalation, Xe-133 equilibrates across the

Xenon-133
The only radioactive gas available commercially is Xe-133.
The physical characteristics of Xe-133 are listed in
Table 15-1. Xe-133 offers superior transit to the periph-
ery of the lung and is very sensitive for detection of
COPD. The half-life of 5.27 days makes for a long shelf-
life. It is available from the radiopharmacy in single- and
multiple-dose vials.
The study is acquired in three phases: an initial single
breath hold, equilibrium, and then washout. Xe-133 rap-
idly clears from the body with a biological half-life of
30 seconds during washout. Therefore, only one view
( usually the posterior) is reliably obtained using a single-
detector system. Although newer dual-head detector
gamma cameras allow the addition of the anterior view, it
is not possible to acquire the multiple views to corre-
spond each position of the perfusion study.
alveolar-capillary membrane and distributes within the
body; however, it clears from the blood as it is exhaled.
However, because Xe-133 is fat soluble, the radiotracer
may accumulate and be retained in the liver of patients
with fatty infiltration (Fig.15-2).
There are some disadvantages to Xe-133. The low
81-keV photopeak is not optimal for the gamma camera
and results in relatively low-resolution images. Ideally, ven-
tilation images would be performed after the perfusion
images in the projection most likely to give diagnostic
information. However, because of its low photopeak,
Xe-133 must be acquired before the Tc-99m MAA perfu-
sion study. If performed after Tc-99m MAA, downscatter
from Tc-99m would degrade the Xe-133 images and make
interpretation difficult.
Being a heavy gas, Xe-133 that escapes into the room
will settle out on the floor. Thus, for radiation safety, it is

Figure 15-2 Xe-133 accumulation in the liver. Posterior ventilation images show delayed
washout of xenon in the lung bases and significant xenon uptake in the region of the liver (arrow,
bottom right image).

mandatory that the room have good airflow and safe exter-
nal ventilation, and must be under negative pressure. On
exhalation, the Xe-133 is directed into tubing that goes to
a xeon “trap,” a charcoal filter that absorbs the expired
xenon and retains it until decay.
Xenon-127
The major advantage of Xe-127 is its higher photopeaks of
172 keV, 203 keV, and 375 keV. These energies are higher
than that of the Tc-99m MAA perfusion agent,so the venti-
lation study can be done after the perfusion exam.
Therefore, only patients with abnormal perfusion scans
would be imaged and positioning could be optimized.
However, the long physical half-life of 36.4 days poses
radiation safety hazards, requiring a different xenon trap
and delivery system.It is not available in the United States.
Krypton-81m
Kr-81m has a photopeak of 190 keV, which permits venti-
lation imaging in multiple projections following the
Tc-99m MAA perfusion study. With a very short half-life
of 13 seconds, Kr-81m must be continuously eluted from
a rubidium-81/krypton-81m generator. Because of this
short half-life, a true equilibrium is not achieved.
Therefore, incomplete penetration of the alveolar spaces
may limit sensitivity for the detection of COPD. The half-
life of the parent, rubidium-81, is only 4.6 hours and the
generator system must be replaced daily. Therefore, this
expensive agent is not practical for most facilities and
Kr-81 has never achieved widespread clinical use.
Radioaerosols
As an alternative to radioactive gases, various aerosolized
particles have been used. The radioaerosol distribution
depicts ventilation during the inhalation phase. The
inhaled aerosol is deposited on the lining of the bron-
choalveolar spaces. Subsequent imaging shows regional
patterns of ventilation. An advantage of the aerosol tech-
nique is the ability to image in multiple views following
a single dose of radiotracer because radioaerosols do not
wash out rapidly like the radioactive gases.
Tc-99m DTPA
The only aerosol approved by the U.S. Food and Drug
Administration ( FDA) is Tc-99m DTPA. Commercial neb-
ulizers are available that provide particles of the appro-
priate size. The ideal aerosol particle size is in the range
of 0.1–0.5 μm. Particles greater than 2–3 μm tend to set-
tle out in large airways including trachea and bronchi.
This central airway clumping may obscure the alveolar
distribution in adjacent lung and even affect the perfu-
sion study performed later by “shining through.”
Clumping most commonly occurs in conditions such as
asthma, COPD, or in patients unable to cooperate with
deep breathing. Although this still happens with Tc-99m
DTPA, it is less problematic than in the past due to tech-
nical improvements, including smaller particle size and
better delivery systems. The aerosol particles cross the
Pulmonary System 513
membrane with a half-life of approximately 45 minutes
and are cleared through the kidneys. This long residence
in the lungs permits imaging in multiple views similar to
the perfusion study.
Technegas
Another agent available in Europe is Technegas,Tc-99m
labeled carbon particles. A Technegas generator pro-
duces very small aerosol particles (0.0005 μm) by
pertechnetate combustion in argon gas. These small par-
ticles do not have the same problem of settling out and
clumping in cases of turbulent flow. Technegas appears
to be superior in image quality to Tc-99m DTPA but is not
available in the United States.
Dosimetry
Radiation dosimetry values are listed in Table 15-2. The
ventilation-perfusion lung scan will result in a low
radiation exposure with either Tc-99m DTPA or Xe-133 as
the ventilation agent. A 5-mCi (185 MBq) dose of Tc-99m
MAA delivers just over 1 rad (1 cGy) to the lungs.
Technique
Ventilation studies are performed before the perfusion
portion of the exam. The higher photopeak of Tc-99m
MAA would cause downscatter into Xe-133 ventilation
images if the perfusion exam was acquired first.
Therefore, Xe-133 imaging is performed first. In the
case of the radioaerosol Tc-99m DTPA, the radiolabel is
the same as the perfusion agent,Tc-99m MAA. For both
studies to be done on the same day without “crosstalk”
interference with each other, the doses must be
adjusted so that the second study overpowers the first
(ratio of at least 3:1). The count rates achievable with
the Tc-99m DTPA nebulizer are much lower than those
obtained on the perfusion images. If the perfusion
study was acquired first, the Tc-99m MAA dose would
need to be limited. The very low dose, on the order of
1 mCi, might not result in optimal diagnostic images.
Table 15-2 Radiation Dosimetry for Ventilation-
Perfusion Radiopharmaceuticals in
rads/mCi (cGy/37 MBq)
Organ Tc-99m MAA Tc-99m DTPA Xe-133
Lung 0.22 0.063 0.0083
Trachea 0.64
Ovaries 0.007 0.012 0.001
Testes 0.008 0.008 0.001
Bladder 0.06 0.17
Whole body 0.015 0.14 0.001
Bold type indicates critical organ.
514 NUCLEAR MEDICINE: THE REQUISITES
Therefore, the ventilation study is obtained first after
inhalation of 800 μCi.
Xenon-133 Ventilation
The protocol for Xe-133 is detailed in Box 15-2. A high-
quality ventilation scan requires patient cooperation.
Because ventilation should be optimized, patients with
clinical bronchospasm should have bronchodilation ther-
apy prior to a lung scan. The study is performed in three
phases: the single-breath or wash-in phase, the equilib-
rium phase, and the washout phase. Severely tachy-
pneic, uncooperative, or unresponsive patients may
require protocol modification; otherwise, ventilation
imaging may not be possible.
Because the location of any perfusion defects cannot
be predicted, the patient is usually positioned for a poste-
rior view. Some laboratories routinely perform both
parts of the study with the patient supine. Others favor
a sitting position if the patient can tolerate it. The sitting
position is generally better because it permits a fuller
excursion of the diaphragm and makes oblique views
easier to obtain during the washout phase. Also, as chest
radiographs are usually obtained in an upright position,
Box 15-2 Protocol for Xenon-133
Ventilation Scintigraphy
PATIENT PREPARATION
None
DOSAGE AND ADMINISTRATION
10–20 mCi (370–740 MBq) inhaled
PROCEDURE
Collimator: Low-energy parallel hole
Photopeak: 20% window centered at 81 keV
Positioning: Patient seated (if possible)
Camera centered over chest posteriorly
If dual head camera, second head may be placed
anteriorly
Use tight seal face mask or mouthpiece with attached
spirometer and intake and exhaust tubing
Acquisition:
First breath: patient exhales fully and is asked to
maximally inspire and hold it long enough (if
possible) to obtain 100,000 counts or 10–15 seconds
Equilibrium: obtain two sequential 90-second images
while the patient breathes normally
Washout:
Turn system to exhaust
Obtain three sequential 45-second posterior images and
then right and left posterior oblique images and final
posterior image
the best correlative information comes from comparison
in similar positions.
To begin the study, a closed system must be set up
with a face mask or mouthpiece. This may be difficult
for some patients to tolerate. The patient is asked to
breathe in and hold a single maximal inspiration of the
Xe-133. An initial image is obtained for 100,000 counts,
if possible, or 10–15 seconds.
The next phase of the study is the equilibrium phase.
After the initial breath image is completed, the patient
breathes a mixture of air and xenon at tidal volume. Two
images are usually obtained for 90 seconds each. In the
third phase, the patient breathes room air and the xenon
is delivered to a trapping system. Three or four sequen-
tial 45-second washout images are obtained. Optional
45-degree posterior oblique images are performed.
Tc-99m DTPA
The lung ventilation protocol using Tc-99m DTPA is listed
in Box 15-3. For studies with the radioaerosol Tc-99m
DTPA, the radiopharmaceutical is placed in a special nebu-
lizer system. The nose is clamped and the patient is asked
to breathe through the delivery system mouthpiece until
sufficient radioaerosol is delivered to the lungs. This may
require several minutes. During this time, the patient may
be placed in a supine position to decrease the apex to base
gradient, but an erect position will improve detection of
Box 15-3 Protocol for Tc-99m DTPA
Ventilation Scintigraphy
PATIENT PREPARATION
None
DOSAGE AND ADMINISTRATION
30 mCi (1110 MBq) Tc-99m DTPA in nebulizer
PROCEDURE
Collimator: Low-energy parallel hole
Photopeak: 20% window centered at 140 keV
Positioning: Place nose clamps on patient and
connect mouthpiece with patient
supine.
Center camera over chest posteriorly
Patient breathes continuously through
mouthpiece for several minutes
Acquisition: Acquire posterior image for 250k and
mark time
Obtain other views for same time
Views: Posterior, anterior, right/left
lateral, right/left posterior oblique,
anterior oblique views recommended

defects in the bases. Although 30 mCi (1110 MBq) of
Tc-99m DTPA is placed in the nebulizer, only 0.5–1.0 mCi
(17.5–37 MBq) is delivered to the patient. The goal is to
deliver enough radioaerosol to the lung so that 200,000- to
250,000-count (200k–250k) images may be obtained in 1–2
minutes. Usually, inhalation continues until 1 mCi (37
MBq) of activity is in the lungs.
The views obtained in the radioaerosol study should
be the same as those obtained in the perfusion phase.
Most nuclear medicine clinics obtain anterior, posterior,
right and left lateral, both posterior 45-degree oblique
views. Right and left 45-degree anterior oblique views
are frequently done and are especially helpful for visual-
izing the lingual segment and medial segment of the right
middle lobe.
Tc-99m MAA
The perfusion study follows the Xe-133 or Tc-99m DTPA
ventilation study. The protocol is described in Box 15-4.
Immediately before injection, the syringe is inverted to
mix the particles as they tend to rapidly settle out and
may clump. The needle should be 23 gauge or larger to
prevent fragmentation of the dose during adminis-
tration. The 2- to 5-mCi (74–185 MBq) dose containing
200,000–500,000 particles is injected over the course of
several respiratory cycles. Care must be taken not to
draw blood back into the syringe. When this happens,
Box 15-4 Protocol for Tc-99m MAA
Perfusion Scintigraphy
PATIENT PREPARATION
None
PATIENT PRECAUTIONS
Pulmonary hypertension: Reduce number of particles
to 100,000
Pregnant patients:Adjust dose lower and observe
requirement for a minimum of 100,000 particles
Right to left cardiac shunt: relative contraindication,
reduce particle number
DOSAGE AND ADMINISTRATION
2–5 mCi (74–185 MBq) I.V. over several respiratory
cycles with patient supine
PROCEDURE
Collimator: Low-energy parallel hole
Photopeak: 20% window centered at 140 keV
Obtain 500k to 750k counts/image
Obtain anterior, posterior, right and left lateral, right and
left posterior oblique
Right and left anterior oblique images optional but
recommended
Pulmonary System 515
clots form in the syringe producing hot emboli on the
lung images ( Fig. 15-3). The patient is in a supine posi-
tion to minimize the gravitational gradient from apex to
base.
Once the injection is complete, imaging can begin
immediately. The patient’s position for imaging should
be the same as that selected for the ventilation portion of
the study. The posterior view is usually acquired for
500,000 counts, and the time noted. The remaining
images can then be set for that amount of time. The
fewest counts will be obtained on the lateral views, as
counts are essentially coming from only the one lung
next to the camera.
Image Interpretation
Perfusion Scintigram
Normal perfusion images should show homogeneous,
uniform distribution of radiotracer throughout the lungs
( Figs. 15-1B and 15-4B). The hilar structures are fre-
quently seen as photopenic areas. The heart is a pho-
topenic defect in the left base on the anterior view and
more mildly decreased activity in the posterior lung.
The spine and sternum attenuate activity along the mid-
line. The pulmonary outline on perfusion images com-
monly appears slightly smaller than on the ventilation
images. This is because of lower spatial resolution on
the Xe-133 ventilation study due to lower counts.
Differences in patient positioning during radiotracer
administration will cause variations in distribution due to
gravity. Pleural effusions will layer out in supine patients
( Fig. 15-5). It is important to know what the positioning
was during imaging to assess defects from breast and
soft-tissue attenuation or from an arm in the field of view.
Attenuation artifacts are also common from metal such as
a pacemaker. Other explanations should be sought for
perfusion defects by examining the radiograph and venti-
lation images.
Small perfusion defects are frequently seen due to var-
ious etiologies. These may be found incidentally and are
more common in smokers. Pulmonary emboli typically
cause multiple moderate or large defects, most com-
monly in the lower lung zones. Causes for perfusion
defects are listed in Box 15-5.
At times, it is necessary to determine if activity outside
the lungs, such as kidney activity, is due to free Tc-99m
pertechnetate or a cardiac shunt. Renal activity can be
seen in both conditions or may occasionally persist fol-
lowing normal systemic absorption of technetium from
the ventilation study. Free pertechnetate in the radio-
pharmaceutical preparation will also be seen as activity
in thyroid and stomach. However,Tc-99m pertechnetate
and other contaminants do not cross the blood–brain
barrier or localize in the brain. In the case of a right-to-
left shunt,Tc-99m MAA lodges in the cerebral capillary
Figure 15-3 Injected blood clot artifact with Tc-99m MAA. Blood clots formed from drawing
blood back into the syringe appear as focal hot spots when they are reinjected into the patient.
These have a variable appearance but can be quite large.

Figure 15-4 Xenon-133 ventilation study. A, The initial breath and equilibrium images are in the
upper row. The sequential washout images in the middle and lower rows show no evidence of air
trapping.
Continued
 Pulmonary System 517

Figure 15-4, cont’d B, Corresponding Tc-99m MAA images show homogeneous distribution of
tracer activity throughout the lungs. C, The chest radiograph was also normal.

circulation, and cortical uptake is diagnostic of a right-to-
left cardiac shunt ( Fig. 15-6). Uptake in the liver indi-
cates colloidal impurities.
Ventilation Scintigram
The normal Xe-133 scan shows homogeneous radiotracer
distribution during all three phases of the study (Fig.15-4A).
The Xe-133 initial breath or wash-in image may show
slightly less uptake compared to the equilibrium rebreath-
ing image. The heart may be seen as a relative photopenic
area in the left lung base. Washout is normally rapid with
a clearance half-time of 2 minutes or less. The last
washout image should have faint or no discernable activ-
ity. In an otherwise normal subject, washout may appear
delayed as a result of the subject’s inability to breathe com-
fortably through the apparatus. Occasionally, abnormal
518 NUCLEAR MEDICINE: THE REQUISITES

Figure 15-5 Pleural effusion effect. A, A-P chest radiograph reveals uniformly greater density in
the right lung compared to the left caused by an effusion layering out posteriorly when the patient
is supine. B, Corresponding Tc-99m MAA perfusion study shows decreased perfusion to the right
lung on the posterior view (upper left hand image), which is not seen on the other views, tipping
off the observer to the explanation for the discrepancy. The pacemaker causes a well-defined
defect (arrow).

activity is seen in the right upper quadrant. This occurs
because xenon is fat soluble and accumulates in the liver
in patients with fatty infiltration of the liver.
The normal distribution of the radioaerosol Tc-99m
DTPA is similar to the initial breath image of a Xe-133
study. When patients are positioned carefully,it is easy to
compare these ventilation images to the multiple projec-
tions from the perfusion study. Activity may be seen in
the mouth and large airways, and swallowed activity may
be seen in the trachea and stomach ( Fig. 15-7).
Differential Diagnosis of Ventilation
Abnormalities
The abnormal ventilation pattern varies depending
on the disease process and the agent used. Abnormal
Tc-99m DTPA images can show areas of absent or
 Pulmonary System 519

decreased uptake. Xe-133 may be abnormal on any one

Box 15-5 Causes of Perfusion Defects or all three phases. The study should be interpreted as
abnormal even if only one phase is abnormal. Many
PRIMARY VASCULAR LESION processes result in air trapping on the Xe-133 scan.
Pulmonary thromboembolism Asthma and COPD with bronchospasm both result in
Septic, fat, and air emboli acutely decreased ventilation. The decreased ventila-
Pulmonary artery hypoplasia or atresia tion, in turn, causes a reflex decrease in perfusion. In
Vasculitis bullae, chronic bronchitis, and bronchiectasis, there is
actual destruction of the bronchial walls with decreased
PRIMARY VENTILATION ABNORMALITY
perfusion in the affected area. Correspondingly, ventila-
Pneumonia tion is decreased with delayed wash-in and air trapping
Atelectasis with delayed clearance.
Pulmonary edema
Although obstructive airway diseases, such as asthma,
Asthma
cause multiple matched defects, pulmonary hyperten-
COPD, emphysema, chronic bronchitis
Bullae sion and restrictive airway disease can cause mismatched
defects. However, the typical pattern of smaller defects
MASS EFFECT in the bases can usually be differentiated from that of PE,
Tumor which causes multiple larger defects.
Adenopathy Occasionally, perfusion deficits involve predomi-
Pleural effusion nantly one lung ( Box 15-6). Although very severe asym-
metric defects are often iatrogenic, other etiologies may
IATROGENIC
be present. A mucous plug usually results in a matched
Surgery—pneumonectomy, lobectomy defect sometimes involving a whole lobe or lung. A cen-
Radiation fibrosis (also postinflammatory fibrosis)
trally located bronchogenic carcinoma or hilar mass can
also cause perfusion defects, which may be matched or

Figure 15-6 Right-to-left cardiac shunt. Tc-99m MAA images reveal abnormal uptake in the
kidneys and brain from a right-to-left shunt. Free pertechnetate activity is seen in the thyroid and
salivary glands but does not explain the cerebral activity.
520 NUCLEAR MEDICINE: THE REQUISITES
Figure 15-7 Swallowed Tc-99m DTPA. Intense uptake
in the trachea and stomach may result from swallowed
radiopharmaceutical.
mismatched ( Fig 15-8). Although a large saddle embolus
could result in a unilateral decrease in perfusion,it is very
uncommon.
Often, the etiology of a ventilation and perfusion
defect is found on the radiograph. Atelectasis, pneumo-
nia,and edema from congestive heart failure are common
examples. Radiographic infiltrates may be seen in PE
with infarct, but pneumonia is a more common cause.
Image Interpretation for Pulmonary Embolism
Diagnosis
A special set of terms has been developed for the diagnos-
tic schemes employed in the interpretation of V/Q scans
( Box 15-7). As described in the section on physiology,
Figure 15-8 Effects of airway obstruction:Ventilation abnormalities may be due to obstructions
in larger airways. This might present as a large defect (left ) caused by bronchogenic carcinoma
or mucous plugs. Constriction of smaller bronchi in asthma can also cause ventilation
abnormalities (right).
Box 15-6 Causes of Severe Unilateral
Lung Perfusion Abnormality
Pneumonectomy
Mediastinal fibrosis
Tumor
Pneumothorax
Mucous plug
Pulmonary embolus
Pulmonary artery stenosis or atresia
Swyer-James syndrome
Chylothorax/massive pleural effusion
perfusion defects may be matched or mismatched depend-
ing on whether or not there is a corresponding ventilatory
abnormality. If Xe-133 is used, the ventilation defect may
not be seen on all phases of the exam, and a defect on any
phase can count as abnormal. The washout phase is most
sensitive for ventilatory abnormalities and detects over
90% of defects compared to 70% for the initial single-
breath image and 20% on the equilibrium view.
The distinction on whether a perfusion defect is
matched or not is fundamental to V/Q interpretation.
Usually, matched defects are due to nonembolic causes.
Acute PE classically results in V/Q mismatch. The embo-
lus blocks blood flow causing a perfusion defect, but the
ventilation remains normal because the airway has no
corresponding blockage. The likelihood of PE increases
as the number of mismatched defects increases, and the
greater the number of defects.
The next important concept is the difference between
a segmental and nonsegmental defect. Perfusion defects
caused by blockage of the pulmonary arterial tree should
reflect the branching or arborization of the pulmonary

Box 15-7 Ventilation-Perfusion
Terminology
V/Q matched defect: Both scans abnormal in same area
and of equal size
V/Q mismatch:Abnormal perfusion in an area of normal
ventilation or a much larger perfusion defect than
ventilation abnormality
Triple-match:V/Q matching defects in a region of chest
x-ray abnormality where the x-ray abnormality is of
the same size or smaller than the radiographic lesion
Segmental defect: Characteristically wedge shaped and
pleural based, conforms to segmental anatomy of the
lung. May be caused by occlusion of pulmonary artery
branches
Large: >75% of a lung segment
Moderate: 25–75% of a lung segment
Small: <25% of a lung segment
Nonsegmental defect: Does not conform to segmental
anatomy or does not appear wedge shaped
circulation in its classic segmental pattern ( Fig. 15-9).
Thus, a classic segmental defect corresponds to one or
more bronchopulmonary segments, is wedge-shaped, and
is pleural-based. Knowledge of segmental anatomy is crit-
ical for correct interpretation. Keeping a diagram at
hand for reference is useful for interpreting V/Q studies
(Fig.15-10). The term nonsegmental is reserved for abnor-
malities due to the patient’s anatomy and those defects
that do not correspond to the pulmonary segments, are
not pleural-based, and do not have the classic wedge
shape. Many conditions resulting in nonsegmental defects
are apparent radiographically, such as pleural effusion,
pneumonia, edema, and tumors. In other cases, it may be
unclear if the defect could be caused by PE. Causes of
nonsegmental defects are summarized in Box 15-8.
Figure 15-9 Effect of pulmonary artery branching pattern
on the appearance of emboli. Emboli may be due to larger, more
proximal clots (left) or showers of smaller clots lodging more
distally (right). In either case, the resulting defects should be
pleural-based and corresponding to the segments of the lung.
Pulmonary System 521
Box 15-8 Causes of Nonsegmental
Defects
Pacemaker artifact
Tumors
Pleural effusion
Trauma
Hemorrhage
Bullae
Cardiomegaly
Hilar adenopathy
Atelectasis
Pneumonia
Aortic ectasia or aneurysm
Assessment of the size of a given defect and determi-
nation of the number of defects present in each cate-
gory are important for the application of the clinical
diagnostic schemes. By convention, a defect is consid-
ered large if it equals more than 75% of the lung seg-
ment, moderate if it is between 25–75% of the size of
a lung segment, and small if it is less than 25% of the
segment. This may be difficult given the variable size of
the different segments. Judgment is subjective and con-
fidence increases with experience. Some authors sug-
gest that there is no significant difference between
a moderate defect and a large one. If this approach is
used, interpretation is simplified.
Evaluation of the chest x-ray is essential. The chest
x-ray findings can alter the final interpretation. The
concept of “matching”also applies to a comparison of the
perfusion exam and the chest radiograph. If a chest x-ray
abnormality is in the area of a perfusion defect, it must be
determined if the finding is acute or chronic. Acute radio-
graphic abnormalities that could be associated with a PE
include atelectasis, infiltrate, or effusion. A defect related
to scar or tumor is not consistent with PE.
PIOPED CRITERIA
Rather than simply calling a scan positive or negative for
pulmonary embolus, risk probability categories have been
developed.If no perfusion abnormalities are demonstrated,
the study is considered normal. Abnormal exams are
interpreted depending on the size and number of perfu-
sion abnormalities, as well as whether they are matched or
mismatched by the ventilation study and chest radiograph,
abnormal studies are categorized as either very low proba-
bility, low probability, intermediate probability, or high
probability. Some physicians will use the term indetermi-
nate interchangeably with intermediate, although most
reserve the term indeterminate for nondiagnostic scans.
522 NUCLEAR MEDICINE: THE REQUISITES

Figure 15-10 Segmental anatomy of the lungs. Upper lobe: 1, apical; 2, posterior; 3, anterior.
Right middle lobe: 4, lateral; 5, medial. Lower lobe: 6, superior; 7, medial basal; 8, anterior basal;
9, lateral basal; 10, posterior basal. Lingula (left): 11, superior lingual; 12, inferior lingual.

The criteria for each of these categories have changed
to some extent over time, and several diagnostic schemes
have been proposed over the years. The Prospective
Investigation of Pulmonary Embolism Diagnosis (PIOPED)
study was a multiinstitutional study sponsored by the
National Institutes of Health to assess the accuracy of
V/Q scintigraphy. Patients underwent V/Q scan, chest
radiograph, and pulmonary angiography. Analysis of the
PIOPED study data resulted in only a few modifications
to prior probability categories. Overall, this prospective
study confirmed much from the published retrospective
studies. The criteria listed in Box 15-9 are a summary of
the widely used modified PIOPED criteria. Further modifi-
cations to the modified PIOPED criteria have been recom-
mended by the PIOPED Nuclear Medicine Working
Group ( Box 15-10). However, data supporting these
changes is more limited.
Normal
When the perfusion study is completely normal, the like-
lihood of PE is less than 5%. The likelihood of significant
morbidity or mortality from PE is less than 1% ( Figs. 15-1
and 15-4).
 Pulmonary System 523

Box 15-9 Modified PIOPED Criteria for High Probability

Pulmonary Embolus Diagnosis
A high-probability interpretation requires identification
of two or more large-segment perfusion defects with
HIGH PROBABILITY SCAN normal ventilation ( mismatched) and a clear chest x-ray.
2 or more large mismatched segmental defects or These perfusion defects can also consist of a combina-
equivalentin moderate/large defects with normal x-ray tion of defects that add up to two or more large defects,
Any perfusion defect substantially larger than if at least one is moderate or large. For example, this
radiographicabnormality might be one large and two moderate defects or four
moderate defects. The perfusion defects must corre-
INTERMEDIATE PROBABILITY SCAN
spond to a bronchopulmonary segment. The high-prob-
Multiple perfusion defects with associated
ability category indicates the risk of pulmonary embolus
x-ray opacities
is greater than 80%. The presence of a matching acute
Greater than 25% of a segment and less than 2
mismatched segmental perfusion defects with radiographic abnormality lowers the likelihood that PE
normal radiograph is present and the study is placed in an intermediate
One moderate segmental probability category ( Figs. 15-11 and 15-12).
One large or two moderate segmental
One large and one moderate segmental
Three moderate segmental Low/Very-Low Probability
Solitary moderate-large matching segmental defect with The chance of a pulmonary embolus in a low-probability
matching x-ray (triple match) scan is less than 20%. A very-low probability category has
Difficult to characterize as high-probability or been introduced with an estimated risk of PE at less than
low-probability
10%. The updated PIOPED recommendations are largely
LOW PROBABILITY SCAN centered on the very-low category, although actual use of
Nonsegmental defects—small effusion blunting the category in practice varies widely. In addition, these
costophrenicangle,cardiomegaly,elevated diaphragm, changes have resulted in a complicated list. The reader
ectatic aorta should start with the basics. It is helpful to remember
Any perfusion defect with substantially that low-probability exams involve small defects and
larger radiographic abnormality matched defects ( Fig. 15-13). Once an exam is placed in
Matched ventilation and perfusion defects with normal the low-probability group, the reading can be refined to
chest radiograph determine if very-low probability criteria apply.
Small subsegmental perfusion defects Whether defects are small or large, or single or multi-
NORMAL SCAN
ple, they can be called low probability. The PIOPED data
found that as long as there was some perfusion in a lung
No perfusion defects
field, even very extensive ventilation-perfusion matched
abnormalities are low probability ( Fig. 15-14). Any num-
ber of perfusion defects can be classified as low probabil-
ity if they are matched by ventilation abnormalities and
the chest x-ray is clear. If all the perfusion defects are
small subsegmental ones (less than 25% of a segment),the
exam is low probability without regard to what the venti-
Box 15-10 Proposed Criteria for Very lation images show if the chest x-ray is clear. However, if
Low Probability there are only one to three such small defects, the study
Interpretation can be called very-low probability. The very-low proba-
bility category also includes nonsegmental defects that
Nonsegmental perfusion defect can be attributed to structures such as the heart, vessels,
Perfusion defect smaller than radiographic lesion or elevated diaphragm on the chest radiograph,
Two or more regional matched defects with normal When the x-ray contains abnormalities, interpretation
x-ray becomes more complex. Any perfusion defect with
One to three small segmental perfusion defects a substantially larger chest x-ray abnormality is also low
Solitary triple-match in upper lung zone or confined to
probability. One way of thinking about this is to consider
single segment
an infiltrate from an infarcted region of lung. The
Stripe sign around perfusion defect on best tangential
view infiltrate should only be seen in areas affected by
the blockage. If the infiltrate is in areas that are being
(continued on p.528)
524 NUCLEAR MEDICINE: THE REQUISITES

LPO Posterior RPO

RAO Anterior LAO

B
Figure 15-11 High probability V/Q. A, Normal posterior Xe-133 images. B, Corresponding Tc-99m
MAA perfusion study reveals multiple bilateral large pleural-based mismatched defects. This pattern
fits the high-probability diagnosis category.
 Pulmonary System 525

Figure 15-12 High-probability V/Q. A, Tc-99m MAA perfusion studies reveals absent perfusion
to the left lower lobe and lingula as well as a moderate sized segmental defect in the right base and a
small one in the right middle lobe. B, Corresponding chest x-ray was normal. C, Pulmonary
arteriogram with injection of the main pulmonary artery shows the large left-sided clot.Some contrast
is seen distal to the clot which does not completely occlude the involved arteries in this case.
526 NUCLEAR MEDICINE: THE REQUISITES

Figure 15-13 Low probability V/Q. A, Xe-133

ventilation images are normal on initial single breath and
equilibrium images (top row) but bibasilar air trapping is
seen on washout (bottom row). B, Tc-99m MAA shows a
large area of relatively decreased perfusion in the left
lower lobe (arrow) and a smaller area in the right base in
the general area of ventilation abnormalities. C,
The radiograph showed no abnormalities in these
matched ventilation perfusion abnormalities.
 Pulmonary System 527

Figure 15-14 Low-probability V/Q. A, Tc-99m DTPA ventilation images reveal severely
decreased ventilation to the right lung with essentially absent ventilation to the left lung. Significant
clumping is present. B, Perfusion images show matching defects, although the abnormalities are
less severe than the ventilation defects on the right. The clumping of the ventilation agent is seen
faintly persisting.
528 NUCLEAR MEDICINE: THE REQUISITES
perfused, it must be due to some other process such as
pneumonia. On the other hand, a perfusion defect that is
much larger than the x-ray abnormality is high probability.
Another case where the abnormal radiograph is
important is the case of the so-called triple match. This
is a segmental perfusion defect which is matched by ven-
tilation and radiographic abnormalities. Although this
combination has traditionally been placed in the interme-
diate category, such abnormalities in the upper lung
fields are probably better categorized as low probability
(Fig. 15-15).
Intermediate Probability
An abnormal study should be placed in the intermedi-
ate category if it does not meet the criteria for high
or low/very-low probability groups ( Fig. 15-16). A sin-
gle moderate-to-large mismatched ventilation-perfusion
defect is considered intermediate probability (Fig. 15-17).
Also, when a matched ventilation-perfusion defect corre-
sponds in size and shape to a radiographic abnormality
(the so-called triple match), it is included in this category
when located in the lower lobes.
The combinations of diagnostic findings in the inter-
mediate category are too numerous to list. Essentially, if
it cannot be placed in the low- or high-probability cate-
gories, it is intermediate probability. Theoretically, the
risk of PE in the intermediate category could be any-
where between 20% to less than 80%.The actual inci-
dence of PE in the intermediate probability category has
been found to be 30–35% in retrospective and prospec-
tive studies.
Pleural effusions are a potential cause of isolated
defects. How large and small effusions are categorized
has changed over recent years and is still somewhat con-
troversial. Some of the confusion is caused by the fact
that PE often causes small effusions. Large effusions are
rare in PE. However, large effusions obscure segmental
perfusion. One report suggests that small effusions with-
out other defects are low probability, and large effusions
can be placed in the very-low probability category.
However,the data is limited. Although small effusions are
generally considered low probability, many experienced
readers still place a large effusion in the low or intermedi-
ate category. It should be noted that a pleural effusion in
PE is rarely present without other perfusion defects.
Special Signs
A number of special signs have been described to aid in
V/Q scan interpretation. The stripe sign refers to a mar-
gin of radioactivity between a perfusion defect and the
pleural surface of the lung ( Fig. 15-18). Because the
pulmonary circulation branches progressively toward
the pleural surface,most pulmonary emboli result in pleu-
ral-based and wedge-shaped defects. The presence of
activity distal to the defect suggests a parenchymal abnor-
mality such as edema or other fluid collection is the cause
rather than PE, and the exam can be placed in the very-
low category. However,the stripe sign is often difficult to
confirm and may be misleading. For example, if it is only
seen in the lateral view, there may be overlap with normal
lung.
The swinging heart sign refers to unusually large car-
diac defects seen on lateral views when the patient has
been imaged lying down and turned to the right and left
for lateral views. The heart has some mobility within the
chest and may compress a variable amount of tissue.
This may create a confusing, changing pattern of activity
depending on position.
Fluid in the pleural space can be difficult to recog-
nize if the patient is imaged in a supine position. This
fluid may layer out and can create an impression that
one lung is generally hypoperfused if the effusion is
asymmetric or unilateral. When patients are imaged in
an upright position, fluid may collect in the major fis-
sure in a curvilinear pattern called the fissure sign
( Fig. 15-19). Blunting of the costophrenic angle from
pleural fluid is commonly seen and positioning will
alter the appearance of the effusion. Subpulmonic
fluid can be missed if careful correlation with an
upright radiograph is not done.
Approach to Interpreting the Ventilation-
Perfusion Scintigram
A rigorous, systematic approach is needed when inter-
preting a V/Q scan. A summarized method of assimilat-
ing the data and applying the PIOPED criteria is outlined
in Table 15-3. First, a current chest x-ray (within 24
hours) must be reviewed. Any radiographic signs of PE
as well as alternative causes for the patient’s symptoms
such as pneumothorax or pneumonia should be sought.
Any chest x-ray abnormality must be classified as acute
or chronic. Acute findings such as atelectasis, infiltrates,
and effusions in the area of perfusion abnormalities that
may cause a triple-match V/Q defect are noted. Chronic
change is unlikely to be related to PE. Cardiomegaly, ele-
vation of the diaphragm, and hilar enlargement should
be noted as they may cause nonsegmental defects.
After examining the chest radiograph, all segmental or
subsegmental perfusion defects are identified on the perfu-
sion scan and recorded by location. Decreased perfusion is
considered abnormal as well as absent perfusion. The ven-
tilation scan is then reviewed in the area of each perfusion
defect. Each of these is, in turn, compared to the radi-
ograph.If the perfusion defect has no radiographic explana-
tion and the ventilation scan is normal, the mismatched
 Pulmonary System 529

Figure 15-15 Triple match V/Q in the upper lobes. A, An AP radiograph raised suspicion for
infarct with a large pleural-based upper-lobe wedge-shaped opacity as well as a left hazy-density from
an effusion layering. B, Perfusion shows a large wedge-shaped perfusion defect in the left upper lobe
and is seen as well as a mild overall decrease on the left corresponding to x-ray abnormalities. C, The
ventilation images show complete matching of the abnormalities on the left. D, A noncontrast CT the
next day also has the same left upper lobe consolidation and effusion. The clinical suspicion was low
in this febrile,immunocompromised patient. The study was interpreted as low probability. The
patient was later found to have aspergillosis not PE.
Figure 15-16 Intermediate probability V/Q. A, Ventilation images with Xe-133 were
unremarkable. B, Perfusion defects are seen in the entire superior segment of the right lower lobe
(arrow) and moderate-sized segmental defect on the left (arrow). C, The chest radiograph
revealed no corresponding abnormality. The combination of one large and one moderate segmental
mismatch places the study in the intermediate probability category.
 Pulmonary System 531

Figure 15-17 Single segmental perfusion defect. Intermediate probability V/Q. A, Perfusion
images demonstrate a large defect in the medial segment of the right middle lobe. B, Ventilation
was unremarkable. C, Close-up views show the utility of the anterior oblique view for visualizing
anterior medial defects.
532 NUCLEAR MEDICINE: THE REQUISITES
Figure 15-18 Stripe sign. Perfusion on the right lateral view
is seen anteriorly along the periphery of the lung beyond an area
of decreased perfusion (arrow), strongly suggesting that the
decreased perfusion in the upper lobe is not due to PE.
sites are candidates for PE. The likelihood of PE increases
with increasing numbers of mismatched defects.
If no moderate or large segmental perfusion defects
are demonstrated, attempts are made to categorize the
study as low probability or less. Matched V/Q defects
with no corresponding chest x-ray abnormalities have
empirically been associated with a low probability of PE.
If a case shows matched abnormalities with a clear chest
x-ray, it is placed in the low probability category when
they are numerous or in the very-low probability if there
are one to three defects. If the perfusion defects are non-
segmental, the study is considered to be in the very-low
probability group.
When the radiograph is abnormal, the size of the
abnormality is compared to the perfusion defect. If
infarcted lung has resulted in an infiltrate, the perfusion
defect is typically as large as or larger than the x-ray
abnormality. Any perfusion defect with a substantially
larger chest radiographic abnormality is in the low
probability category. In practice, size comparisons may
be difficult, and thus these criteria should be used cau-
tiously and sparingly. Cases of triple-matched abnormali-
ties on perfusion, ventilation, and radiographic studies
could be considered intermediate probability.
However, there is limited data suggesting that the
chance that a pulmonary embolus is present also
depends on the location of the defect. Abnormalities
confined to the upper or middle lung zones less likely
to be the result of an acute pulmonary embolus.
Therefore, only triple-matched defects in the lower lung
zones were intermediate probability, whereas such
defects in the upper and middle lung zones are low
probability for PE ( Table 15-4).
If the study cannot be categorized as normal, high
probability or low/very-low probability, then it is placed
in the intermediate probability category. When there are
problems performing the ventilation scan, abnormal per-
fusion studies may be considered indeterminate. Also,
indeterminate studies occur when posterior perfusion
images cannot be obtained on portable studies per-
formed at the patient’s bedside.
Accuracy of Ventilation-Perfusion
Scintigraphy
The agreement between readers for V/Q scan interpreta-
tion varies widely. Interobserver agreement is high for
normal and high probability exams ( 90–95%). However,
less agreement was seen deciding which studies should
be intermediate probability or low probability ( 70–75%).
Adherence to a strict set of criteria may help minimize
variability between readers.
Clinical risk factor scores can be added to the V/Q inter-
pretive process to improve accuracy. When clinicians
had a high degree of suspicion for pulmonary embolus,
the incidence of pulmonary embolus was increased to
96% among patients with a high probability. The clinical
picture also had great impact among low probability
exams. If the pretest suspicion for PE was high, the likeli-
hood of PE was not less than 20% as might be expected
but increased to 40%. On the other hand, when the clini-
cal probability was low and the scintigraphic interpreta-
tion was normal or near normal,less than 2% had PE.
Another important observation from the PIOPED
study is the low likelihood of an adverse clinical outcome
in patients with normal and low probability scintigraphic
patterns. In the study, 150 patients who had either a low
probability or normal/near-normal scan but who did not
undergo angiography were followed for at least 1 year.
No patient had an adverse event or was readmitted for
suspected PE. Some may well have had small pulmonary
emboli, but none received anticoagulation therapy and
the clinical course was unremarkable. This finding sup-
ports several other studies that suggest a benign clinical
course in low probability cases.
In the PIOPED trial, the specificity of a high probabil-
ity V/Q scan was 97% and the positive predictive value
was 88%. However, the sensitivity for pulmonary embo-
lus of a high probability scan is only 41%. This means the
majority of patients with pulmonary emboli have an
intermediate or low probability scan. This sensitivity is
disappointing; if the criteria were relaxed, the sensitivity
for PE would increase, but this would be at the expense
of the specificity ( Table 15-5). The high specificity of
the high probability category allows initiation of antico-
 Pulmonary System 533

Figure 15-19 Fissure sign. A, Tc-99m MAA perfusion images show a curvilinear defect in the
area of the major fissure of the right lung from the “fissure sign.” B, Corresponding radiograph
shows right costophrenic angle blunting but provides no indication of the extent of the fluid in the
fissure.
534 NUCLEAR MEDICINE: THE REQUISITES
agulation therapy in the appropriate clinical situation.
Patients at risk for anticoagulation therapy may still
require angiography.
Differential Diagnosis of V/Q
Mismatches
The high probability category means there is a greater than
80% probability of pulmonary embolus. However, many
processes can cause mismatched defects (Box 15-11). These
can lead to false-positive readings. Of the potential causes of
false-positive interpretations,or so-called PE mimics,one of
the most common is an unresolved PE. Although PE typically
resolves in younger patients,complete resolution is less com-
mon in the elderly. Portions of perfusion defects from pul-
monary emboli may persist and remain mismatched. One
study demonstrated that as many as 35% of acute emboli do
not completely resolve. The positive predictive value of a
high probability reading decreases from 91% in patients with-
out prior PE to 74% in those with a history of prior PE.
The time course of defect resolution varies. After the
acute event, mismatched defects generally persist for
more than 24 hours. Large mismatched defects tend to
decrease over time or fragment into smaller lesions in the
periphery. Although exams may become normal in
hours or days, abnormalities can persist for weeks or
months. In fact, follow-up studies show that defects per-
sisting at 3 months are likely to remain unresolved. After
a high probability V/Q or a positive CTPA for PE, a follow-
up V/Q is useful to establish a new baseline in the event
that symptoms recur ( Fig. 15-20). It is usually recom-
mended that this new baseline is delayed for 3 weeks to
3 months to allow time for resolution.
Causes for mismatched defects other than PE are often
discovered by taking a careful history. A patient may have
a known hilar tumor, which could compress the pul-
monary vessels, particularly the veins, before impacting
the airway. Radiation effects and drug abuse are also com-
mon etiologies. Certain defined clinical situations may
increase the possibility for septic emboli and fat emboli.
COMPUTED TOMOGRAPHY
PULMONARY ARTERIOGRAPHY FOR
THE DIAGNOSIS OF PULMONARY
EMBOLI
Increasingly, CTPA is replacing the V/Q scan for the diag-
nosis of pulmonary emboli. Although the reported sensi-
tivity of CTPA has varied greatly over recent years, from
36% to 100%, and early studies found poor detectability of
subsegmental emboli. However, multidetector technology
and interpretive experience have advanced rapidly.Recent
multi-center data suggest that CTPA has a very high nega-
tive predictive value for pulmonary emboli, in the range of
99% (i.e., a negative study rules out pulmonary emboli). If
a pulmonary vascular filling defect is seen,it is highly accu-
rate for the diagnosis of pulmonary embolus.
The NIH-funded Prospective Investigation of
Pulmonary Embolism II (PIOPED II) was initiated to
determine the accuracy of CTPA for the diagnosis of PE
as well as the added utility of CT venous examination of
the pelvis and thighs. The results of this trial have not
been reported as of this writing.
The V/Q scan has stood the test of time and remains
an important diagnostic tool for the diagnosis of PE. It
will continue to have an important role in patients with
a history of contrast reaction, renal failure, pregnancy,
and inconclusive results on CTPA.
OTHER APPLICATIONS OF
VENTILATION-PERFUSION
SCINTIGRAPHY
Quantitative Lung Scan
Quantification of lung perfusion and ventilation can be
valuable in the preoperative assessment for operability of
high-risk patients prior to planned lung resection for
malignancy, dead-space lung volume reduction in severe
COPD, and lung transplantation. This information is used
in conjunction with respiratory spirometry (e.g., FEV1,
FVC). It can also be useful in assessing relative pulmonary
perfusion before and after operations for congenital heart
disease (e.g., correction of pulmonary stenosis).
Right to left lung differential function is commonly
performed by acquiring the anterior and posterior views,
drawing regions of interest around the right and left
lungs and calculating the geometric mean to correct for
attenuation (Fig. 15-22).
Geometric mean = √ (countsanterior × countsposterior )
However, the anterior and posterior views do not
allow for good separation of the upper and lower lobes.
Posterior oblique views allow clear separation of the
upper and lower lobes (Fig. 15-23).
Adult Respiratory Distress Syndrome
The clearance of Tc-99m DTPA is significantly affected by
the presence of pulmonary disease. The clearance half-
time is approximately 45 minutes in healthy adults.
Patients with adult respiratory distress syndrome have
more rapid clearance, probably due to the rapid diffusion
of Tc-99m DTPA across the airspace epithelium to the pul-
monary circulation. Other conditions associated with
 Pulmonary System 535

Table 15-3 Analysis of Ventilation Perfusion Scans Box 15-11 Conditions Associated with
Ventilation-Perfusion
Perfusion Probability Mismatch
defect(s) Ventilation Chest x-ray category

Acute pulmonary embolus

MODERATE TO LARGE Chronic pulmonary embolus
≥ 2 segments Mismatch Clear High Other causes of embolism (septic, drug abuse,
(or equivalent) iatrogenic, fat)
<2 Mismatch Clear Intermediate Bronchogenic carcinoma (and other tumors)
Match Lower lung Intermediate Mediastinal or hilar adenopathy (with obstruction of
zone pulmonary artery or veins)
abnormality Hypoplasia or aplasia of pulmonary artery
Match Upper lung Low
Swyer-James syndrome (occasional)
zone
abnormality
Postradiation therapy
Multiple Match Clear Low Vasculitis

SMALL (<25% OF A SEGMENT)

>3 N/A Clear Low
1–3 N/A Clear Very low

NONSEGMENTAL
N/A Shows Very low DETECTION OF DEEP VENOUS
anatomical THROMBOSIS
reason for
perfusion Although the major thrust of this chapter is the diagnosis
abnormality
of PE,diagnosis of deep venous thrombi (DVT) is a related
NONE topic. Two tests no longer frequently utilized are tradi-
N/A N/A Normal tional contrast venography and radionuclide venography.
Radionuclide venography involved placing a tourniquet
NA, Not applicable. above the ankle and injecting radiopharmaceutical,
Tc-99m MAA, into a small vein in the foot ( Fig. 15-24).
Although radionuclide venography is very sensitive above
Table 15-4 Positive Predictive Value of Triple-
the knee, interpretation is difficult below the knee.
Matched Perfusion Defects Based on
Doppler sonography imaging in combination with vari-
Location
ous compression techniques is useful in the veins of
the lower extremity above the knee. Contrast CT can
95% confidence
Lung zone Defects from PE (%) interval
be used to assess the pelvic veins. Images of the pelvis
can be obtained after the pulmonary arterial imaging
Upper 11% 3–26% is completed. It may be technically difficult to time
Middle 12% 4–23% the study to achieve adequate venous opacification for
Lower 33% 27–41% diagnosis.
PIOPED II also examined utility of CTV for the pelvic
and thigh veins. Preliminary reports found CTV to be
Table 15-5 V/Q Scan Sensitivity and Specificity 95% sensitive with a positive predictive value of 86%
overall. However, the majority of lesions were found in
Probability category Sensitivity (%) Specificity (%)
the thighs. These would be potentially visualized by
ultrasound without the exposure to ionizing radiation.
High 41 97 Various radiolabeled agents have been developed that
High + intermediate 82 52 bind to components of an active clot. These have included
High + intermediate + low 98 10 radiolabeled platelets, fibrin, and various peptides. Tc-99m
apcitide (AcuTect) has been approved by the FDA for the
diagnosis of DVT (Fig. 15-25). It is useful in patients with
increased Tc-99m DTPA clearance are cigarette smoking, equivocal ultrasound findings or for the detection of throm-
alveolitis, and hyaline membrane disease in infants. This bus in the calf. Tc-99m apcitide binds to a glycoprotein
technique has not found a clear-cut clinical use. receptor (GpIIb–IIIa) in acute DVT.
536 NUCLEAR MEDICINE: THE REQUISITES

The protocol calls for intravenous injection of 20 mCi

( 740 MBq) of Tc-99m apcitide. Images of the calf, thigh,
and pelvis are obtained at 10–15 minutes and 60–90 min-
utes. The reported sensitivity and specificity of Tc-99m
apcitide varies, but is roughly 83% compared to contrast
venography with a specificity of nearly 91%. It is not
approved for the diagnosis of pulmonary emboli. The
use of radiolabeled peptides in the diagnosis of pul-
monary emboli continues to be investigated.

Figure 15-20 Resolution of PE. Normal ventilation images

(left ) and abnormal perfusion images (right) were obtained in the
same patient at different times. A, The symptomatic patient had a
high probability V/Q scan. B, Ten years later, recurrent symptoms
led to another V/Q with marked mismatches. It is difficult to tell
how many of these are new as no follow-up baseline study was
performed after the original episode. C, Near complete resolution
of these defects 7 days later confirms the mismatches were new
and due to recurrent PE.

Figure 15-21 CT pulmonary arteriography. Posterior equilibrium ventilation image (A) and
perfusion image (B) reveal an extensive mismatch in the left lung base and upper half of the right lung.
Continued
 Pulmonary System 537

Figure 15-21, cont’d C, Spiral CT shows contrast around a large partially obscuring clot in the
left main pulmonary artery and right upper lobe pulmonary artery. D, Filling defects in the left
descending pulmonary artery and right interlobar pulmonary artery are also present.

Figure 15-22 Preoperative quantitative lung scan. Anterior and posterior projections. Geometric
mean quantification is performed for left to right lung differential function as well as upper, mid, and
lower lung regions.
538 NUCLEAR MEDICINE: THE REQUISITES

roi1
roi1 roi2
roi0

LPO RPO

Left upper: 23% Right upper: 38%

Left lower: 77% Right lower: 62%
Figure 15-23 Quantitative lung scan with bilateral decreased function in the upper lobes in a
patient with chronic obstructive pulmonary disease.

Figure 15-24 Abnormal radionuclide venogram patterns.

A, Venous obstruction (arrow) on the right caused by thrombus.
B, Extensive collateralization on the right indicates obstruction of Figure 15-25 Acute thrombophlebitis with Tc-99m apcitide.
the deep venous system. Increased uptake is seen in the area of the right calf.

RECOMMENDED READING
Freeman LM, Krynyckyi B, Zuckier LS: Enhance lung scan diag-
nosis of pulmonary embolism with the use of ancillary scinti-
graphic findings and clinical correlation. Semin Nucl Med 31:
143-157, 2001.
Goldberg SN, Richardson DD, Palmer EL, Scott JA: Pleural
effusion and the ventilation-perfusion scan interpretation
for acute pulmonary embolus. J Nucl Med 37:1310–1318,1996.
Gottschalk A, Sostman HD, Coleman RE, et al: Ventilation-
perfusion scintigraphy in the PIOPED study. Part II. Evaluation
of the scintigraphic criteria and interpretation. J Nucl Med
34:1119–1126, 1993.
Gottschalk A, Stein P, Goodman LR, Sostman HD: Overview of
prospective investigation of pulmonary embolism diagnosis II.
Semin Nucl Med 32: 173–182, 2002.
Gray HW: The natural history of venous thromboembolism:
impact on ventilation/perfusion scan reporting. Semin Nucl
Med 32: 159–172, 2002.
Pulmonary System 539
Hatabu H, Uematsu H, Nguyen B: CT and MR in pulmonary
embolism: a changing role for nuclear medicine in diagnostic
strategy. Semin Nucl Med 32: 183–192, 2002.
Quiroz R, Kucher N, Zou KH, et al: Clinical validity of a negative
computed tomography scan in patients with suspected pul-
monary embolism: a systemic review. JAMA 293(16):2012–
2017, 2005.
Stein PD, Henry JW, Gottschalk A: Mismatched vascular defects:
an easy alternative to mismatched segmental equivalent defects
for the interpretation on ventilation/perfusion lung scans in
pulmonary embolism. Chest 104: 1468–1471, 1993.
Worsley DF, Alavi A: Comprehensive analysis of the results of
the PIOPED study. Prospective investigation of pulmonary
embolism diagnosis study. J Nucl Med 36: 2380–2387, 1995.
Worsley DF, Kim CK,Alavia A, Palevsky HI: Detailed analysis of
patients with matched ventilation-perfusion defects and chest
radiographic opacities. J Nucl Med 34:1851–1853, 1993.
 16
CHAPTER Pearls, Pitfalls, and
Frequently Asked
Questions

Radiopharmaceuticals A: The parent radionuclide must have a long enough

Nuclear Medicine Physics
Radiation Detection and Instrumentation
SPECT and PET
Endocrine
Bone
Hepatobiliary
Genitourinary
Oncology—Positron radiopharmaceuticals
Oncology—Single photon radiopharmaceuticals
Gastrointestinal
Infection and Inflammation
Central Nervous System
Cardiac
Pulmonary
This chapter reinforces concepts presented in this book.
Every student of medicine gathers pearls of wisdom from
his or her mentors that may not fit well into a didactic
treatment of a subject but that are extraordinarily valu-
able in day-to-day practice. Likewise, we all learn to avoid
pitfalls that arise in situations but that have escaped our
formal education.Also, questions posed with interpreta-
tion require assembling multiple bits of information for
a correct answer, and these questions never seem to be
presented in quite the same way that subject material
was presented didactically. By its nature, this chapter is
neither comprehensive nor weighted to the relative
importance of the topics.
RADIOPHARMACEUTICALS
Q: What relationship between the half-lives of a parent
radionuclide and a daughter radionuclide is neces-
sary for a generator system?
half-life to permit formulation and distribution of
the generator.The daughter half-life must be reason-
able for clinical application. A longer-lived parent
decays to a shorter-lived daughter in all generator
systems in use.
Q: How are parent and daughter radionuclides sepa-
rated in generator systems?
A: Because the parent and daughter are different ele-
ments, they can be chemically separated.
Q: What is the major drawback of molybdenum-99 pre-
pared by neutron activation?
A: When Mo-99 is prepared from Mo-98 by neutron
activation, the two isotopes cannot be separated,
and significant Mo-98 carrier exists in the prepa-
ration. This ultimately results in low specific con-
centration eluates of technetium-99m from the
generator system.
Q: What is the difference between transient equilib-
rium generators and secular equilibrium generators?
A: In secular equilibrium generators,the half-life of the
parent is far longer than the half-life of the daughter.
If the generator system is left alone,the activity of the
daughter becomes equal to that of the parent. In gen-
erator systems in which the parent half-life is 10 to
100 times that of the daughter, a condition of tran-
sient equilibrium occurs if the generator is not
eluted.The point of transient equilibrium is defined
as the time at which the ratio of the daughter and
parent activities becomes a constant. Because the
parent half-life is longer, the daughter appears to
decay with the same half-life.The Mo-99/Tc-99m gen-
erator system is an example of transient equilibrium.

540

Q: What is the practical problem with having carrier
Tc-99 in the generator eluate?
A: Tc-99 and Tc-99m behave identically from a chemi-
cal standpoint.Therefore, if there is excessive Tc-99
in the eluate, labeling efficiency can be impaired.
For example, in a kit preparation using stannous
chloride as a reducing agent, there may be unre-
duced Tc-99 and Tc-99m left in the preparation, with
the consequent presence of radiochemical impuri-
ties in the final preparation.
Q: When is the buildup of Tc-99 at its highest?
A: Because Tc-99 has a far longer half-life than Tc-99m,
the longer the interval between generator elutions,
the greater the buildup of Tc-99. The first elution
after commercial shipment or after a long weekend
will have the highest content of Tc-99.
Q: What is the legal limit for Mo-99 in Tc-99m-contain-
ing radiopharmaceuticals?
A: The Nuclear Regulatory Commission limit is 0.15
mCi of Mo-99 activity per 1 mCi of Tc-99m activity in
the administered dose.
Q: How does the ratio of Mo-99 to Tc-99m change with
time?
A: In any preparation in which the radionuclidic con-
taminants have longer half-lives than the desired
radionuclide label, the relative activity of the con-
taminant increases with time. This is an issue for
iodine-123 preparations that have longer-lived
radioiodine contaminants, as well as for the Mo-99
contamination in Tc-99m preparations.
Q: What is the purpose of stannous ion in Tc-99m label-
ing procedures?
A: Stannous ion is used to reduce technetium from a +7
valence state in pertechnetate to lower valence
states necessary for labeling a wide range of agents.
The development of this approach was a major
breakthrough in nuclear pharmacy.
Q: What is Webster’s rule in regard to the dose of
a pharmaceutical to administer to a child?
A: It is based on the child’s age: (age + 1)/(age + 7) ×
adult dose.
Q: What constitutes a medical event, formerly known
as a misadministration of a radiopharmaceutical?
Pearls, Pitfalls, and Frequently Asked Questions 541
A: A misadministration was formerly defined by the
NRC as a radiopharmaceutical given to the wrong
patient, a patient receiving the wrong radiopharma-
ceutical, receiving the ordered radiopharmaceutical
by the wrong route of administration, or the adminis-
tered dose differing from the prescribed dose by
greater than an allowable standard.Although these
are all of concern and need discussion of quality
assurance within a department and institution as well
as a record of the event, the NRC now only requires
reporting medical events where the effective dose
equivalent to the patient exceeds 5 rem to the whole
body or 50 rem to an individual organ or a diagnostic
dose of I-131 exceeds 30 μCi.
Q: Describe the general response to the spill of radioac-
tive material.
A: In general, the person who recognizes that a spill
has occurred should notify all persons in the vicin-
ity, and the area should be restricted. If possible, the
spill should be covered. For minor spills, cleanup
using appropriate disposable and protective cloth-
ing can be accomplished until background or near-
background radiation levels are observed. For
major spills, the source of the radioactivity should
be shielded. For both major and minor spills, all
personnel potentially exposed in the area should
be surveyed, with appropriate removal of contami-
nated clothing and decontamination of skin. The
radiation safety officer should be notified of all
spills and has the primary responsibility for super-
vising cleanup for major spills and determining
what reports must be made to regulatory agencies.
NUCLEAR MEDICINE PHYSICS
Pearl
Positrons are positive electrons, thus particles. With
radioactive decay,an emitted positron travels 2–10 mm in
tissue (depending on the radionuclide) before losing its
kinetic energy, then interacting with an electron.The two
particles annihilate each other and emit two 511-keV
gamma photons at approximately 180-degree angles
from each other.The gamma photons can be detected by
positron emission tomography (PET) coincidence
542 NUCLEAR MEDICINE: THE REQUISITES
detectors.This conversion of mass to energy is predicted
by Einstein’s well-known formula: E = mc2.
Q: What is the difference between x-rays and gamma rays?
A: Both x-rays and gamma rays are types of ionizing
radiation. By definition, x-rays originate outside the
atomic nucleus, and gamma rays originate inside the
atomic nucleus.The respective energy spectra for
x-rays and gamma rays substantially overlap at the
high-energy end of the spectrum for all forms of
electromagnetic radiation.
Q: What is the energy equivalent of the rest mass of an
electron?
A: 511 keV. This is also the energy equivalent of
a positron (positive electron).
Q: What is the difference between the rad, roentgen,
and rem?
A: These terms are frequently confused with each other
but have important distinctions. Rad stands for radia-
tion absorbed dose.A rad is equal to the absorption
of 100 ergs per gram of absorbing material.The rad is
the traditional unit of absorbed dose.The gray (Gy) is
the unit of absorbed dose in the International System
of Units (SI). One gray = 100 rads.
Rem is an acronym for roentgen equivalent man.The
rem is calculated by multiplying the absorbed dose in
rads by a factor to correct for the relative biological
effectiveness (RBE) of the type of radiation in question.
The rem is the traditional unit. In the SI system, the
term sievert (Sv) is used.One sievert = 100 rem.
The roentgen (R) is a unit of radiation exposure.It is
defined as the quantity of x-radiation or gamma radia-
tion that produces one electrostatic unit of charge per
cubic centimeter of air at standard temperature and
pressure. In the SI system, radiation exposure is
expressed in terms of coulombs per kilogram (C/kg).
One roentgen is equal to 2.58 ¥ 104 C/kg air.
Q: Which is more penetrating in soft tissues, alpha par-
ticles or beta particles of the same kinetic energy?
A: Alpha particles have very low penetration in soft tis-
sue because of their rapid loss of kinetic energy
through interaction of their electrical charge with
electrons in the tissues. Beta particles of the same
respective kinetic energy of alpha particles have
higher velocity, lower mass, and a single negative
charge.They demonstrate significantly greater pene-
tration in soft tissues, although penetration still is
typically measured in millimeters.
Q: Define the two systems for expressing radioactive
decay.
A: The traditional unit of radioactive decay is the curie
(Ci). One curie is equal to 3.7 × 1010 disintegrations
per second (dps).This number was derived from the
decay rate of 1 gram of radium. (Modern measure-
ments indicate that the actual decay rate for 1 gram
of radium is 3.6 × 1010 dps.) In the SI system, decay
is expressed in becquerels (Bcq). One becquerel
equals one disintegration per second.
Q: How are the half-life and the decay constant related?
A: The physical half-life (T1⁄ 2) of a radionuclide is
defined as the time for half the atoms in a sample to
decay.The half-life is expressed in units of time, typi-
cally seconds, minutes, hours, days, or years. The
decay constant indicates the fraction of the sample
decaying in a unit of time. The units of the decay
constant are “per unit time” (per second, per hour).
Mathematically, the half-life (T1⁄2) and the decay con-
stant (λ) are related by the following equation:
T ⁄ = ln 2
1
λ
2
Q: Which is longer, the biological half-life or the effec-
tive half-life?
A: The effective half-life is always shorter than either the
biological half-life or the physical half-life because bio-
logical clearance and physical decay take place simul-
taneously. In calculation of radiopharmaceutical
dosimetry, the conservative assumption is sometimes
made that the biological half-life is infinite. This is
probably never completely correct but simplifies cal-
culations because the effective half-life may be taken
simplistically as the physical half-life.
Q: After a photon has undergone Compton scattering,
how does the energy of the scattered photon com-
pare to the original photon energy?
A: In Compton scattering, the photon gives up energy
to a recoil or Compton electron.The “scattered”pho-
ton has correspondingly lower energy.The amount
of energy lost increases as the angle of scattering
increases.
Q: What factors speed up or slow down radioactive
decay?
A: Unlike chemical reactions, radioactive decay is
a physical constant that cannot be sped up or
slowed down by heating or cooling a specimen or
by applying other physical or chemical influences.

Q: How many observed counts are necessary to have
a percent fractional standard deviation of 5%, 2%,
and 1%, respectively?
A: 400, 2500, and 10,000 counts, respectively.
Q: What is the maximum number of electrons that can
occupy the outermost shell of an atom?
A: Eight.
Q: What special term is used to designate the electrons
in the outermost shell of an atom?
A: They are called valence electrons and are responsi-
ble for many of the chemical characteristics of the
element.
Q: What is the binding energy of an electron?
A: Binding energy refers to the amount of energy
required to remove that electron from the atom.
Electrons in shells close to the nucleus have higher
binding energy than electrons farther from the
nucleus.This energy is typically expressed in terms
of electron volts (eV). Remember that the binding
energy for each electron shell and subshell is charac-
teristic for the respective element; the higher the
atomic number of the element, the greater the bind-
ing energy for each shell and subshell.
RADIATION DETECTION AND
INSTRUMENTATION
Q: What are some examples of the uses of ionization
chambers in nuclear medicine?
A: Ionization chambers are often used in radiation sur-
vey meters and some pocket dosimeters. The
radionuclide dose calibrator incorporates an ioniza-
tion chamber.
Q: What is the purpose of the thallium impurity added
to sodium iodide crystals?
A: The thallium is used to “activate” the sodium iodide
crystal.The thallium impurity provides “easier” path-
ways for the return of electrons from the conduc-
tion band of the crystal to the valence bands of
atoms.
Q: What is the relationship between photon energy
and detection efficiency in a sodium iodide crystal?
A: For a given crystal size,detection efficiency decreases
with increasing photon energy.
Q: Why do photopeaks appear as bell-shaped curves in
pulse height spectra rather than as discrete spikes
corresponding to the energy of the gamma ray?
Pearls, Pitfalls, and Frequently Asked Questions 543
A: Although gamma rays have discrete energies, the
detection process is subject to statistical factors at
each step of the process.The bell-shaped curve corre-
sponding to the gamma ray photopeak reflects these
statistical variations, which results in different events
being measured as having slightly different energies.
The better the “energy resolution” of a pulse height
analyzer,the narrower the bell-shaped curve.
Q: In using a gamma scintillation camera, what does it
mean to “set”the energy window?
A: Gamma cameras are equipped with pulse height
analyzers that allow the operator to select a range of
observed energies for accepting photons to be used
in making the scintigraphic image.The “window” is
usually described by giving the photopeak energy of
interest and a percentage range that defines the lim-
its of acceptance above and below the photopeak
energy.A typical window for the 140-keV photon of
technetium-99m is 20%, or ±14 keV.
Q: What are the causes of homogeneous flood field
images in gamma camera quality control?
A: Causes include improper photomultiplier tube volt-
age adjustment, off-peak camera pulse height ana-
lyzer setting, crystal imperfections or damage, poor
coupling of the crystal and the photomultiplier
tubes, and inadequate mixing of radioactive tracer in
the flood phantom.
Pitfall
Some nuclear medicine clinics use radioactivity in the
patient to confirm the window setting.This can be a pit-
fall because scattered photons are included in the
observed spectrum and can actually shift the apparent
location of the photopeak. Ideally, a sample of the
radionuclide to be imaged should be used for “peaking”
in the gamma camera energy window.
Q: What effects do Compton-scattered photons have
on scintigraphic image quality?
A: Compton-scattered photons are the enemy! Scattered
photons that fall within the acceptance limits of the
energy window are included in the image.They rep-
resent false data because they are recorded in a dif-
ferent spatial location than the origin of the primary
photon. Thus Compton scattering reduces image
contrast and spatial resolution.Also, Compton-scat-
tered photons falling outside the energy window still
must be processed by the gamma-camera pulse-
height analyzer circuitry.These rejected events con-
tribute to dead time and reduce the count rate
capability of gamma cameras.
544 NUCLEAR MEDICINE: THE REQUISITES
Q: What photons are desired in the scintigraphic
image?
A: Primary (unscattered) photons that arise in the
organ of interest in the body and travel parallel
to the axis of the gamma-camera collimator field-of-
view are the photons desired in the image.
Intuitively,one may think of these as “good”photons.
All other photons are “bad” photons.These include
primary (unscattered) photons that arise in the
object or organ of interest but travel “off axis,” pri-
mary photons that arise in front of or behind the
organ of interest (background photons), and all scat-
tered photons.
Q: What is the purpose of the collimator?
A: The collimator defines the geometric field-of-view
of the gamma-camera crystal. Off-axis photons,
whether they are primary photons or scattered pho-
tons, are absorbed in the septa of the collimator.
Pearl
Pinhole collimators allow resolution of objects below the
spatial resolution of the gamma camera through geomet-
ric magnification.
Q: What is the construction difference between a low-
energy all-purpose collimator and a low-energy
high-resolution collimator?
A: A high-resolution collimator has more holes that are
smaller and deeper.
Q: How does poor energy resolution degrade spatial
resolution?
A: Gamma cameras with poor energy resolution have
reduced ability to reject scattered photons on the
basis of pulse height analysis, as well as reduced abil-
ity for accurate determination of x and y coordinates
for spatial localization of events.
SPECT AND PET
Pearl
Most nuclear medicine departments use 180-degree
SPECT acquisition for cardiac studies and 360 degrees for
imaging other organs, including the brain.
Pearl
For SPECT imaging the highest resolution collimator that
provides sufficient count rate should be selected.
Pitfall
Besides equipment factors, patient motion is the most
important cause of image degradation in SPECT and PET
studies.
Q: What special importance does the biological half-life
of a radiotracer have in SPECT imaging?
A: In SPECT imaging, data are acquired sequentially
from different sampling angles. If significant biologi-
cal redistribution of a radiopharmaceutical takes
place between the start of data acquisition and com-
pletion, the reconstruction of tomographic images
can be significantly distorted.
Q: What is a filter?
A: Filters are special mathematical functions applied to
SPECT and PET data that enhance desired character-
istics in the image, such as background subtraction,
edge enhancement, and suppression of statistical
noise. The ramp filter is designed to eliminate or
reduce the star artifact.
Q: What is the star artifact in SPECT and PET recon-
struction?
A: A star artifact results from simple unfiltered back-
projection of a point source.
Q: What are the two basic approaches to attenuation
correction?
A: The two basic approaches are the analytical or math-
ematical approach and the empirical approach. In
the analytical approach, attenuation correction is
estimated from a model of the body part under
investigation. In the empirical approach, attenuation
correction is accomplished by direct measurement
using transmission scanning.
Pearl
One of the great advantages of SPECT and PET is the abil-
ity to perform flexible reformatting of image data in mul-
tiple image planes. For cardiac imaging, short-axis,
vertical long-axis, and horizontal long-axis views of the
heart are typically obtained.
Pearl
Two quick ways of assessing patient motion during
SPECT imaging are to view the projection images as a cin-
ematic closed-loop display and to create slice sinograms.
In the cinematic display, patient motion is seen as
a flicker from one projection image to another. On sino-
grams, patient motion is seen as a discontinuity in the
stacked projection profiles.

Pitfall
SPECT is subject to a number of artifacts. Field flood
nonuniformity can result in ring artifacts. Center-of-
rotation misalignment causes loss of image resolution
and if severe, ring artifacts.
Pearl
PET imaging relies on the coincidence detection of the
two gamma ray photons given off simultaneously during
a positron annihilation event.
Pitfall
The higher the overall count rate in PET imaging, the
more likely the recording of “false” events owing to
the presence of paired random events that appear to the
detection circuitry as paired annihilation photons.
Pearl
The spatial resolution of PET is twice or more that of
SPECT.
Pitfall
Spatial resolution in PET is limited by positron travel in
soft tissue before annihilation and photon emission.
Pearl
PET imaging with transmission attenuation correction
and detector sensitivity calibration allows absolute quan-
titative uptake determinations.
Pearl
Positron emitters, such as carbon, nitrogen, oxygen, and
fluorine (as replacement for hydrogen) make possible
the potential radiolabeling of any biological compound.
The chemistry for developing and radiolabeling single-
photon radiopharmaceuticals is usually considerably
more complex.
ENDOCRINE
Q: What is the origin of lingual and sublingual thyroid
tissue?
A: The main thyroid anlage begins as a downgrowth
from the foramen cecum.Thyroid tissue may be seen
anywhere along the tract of the thyroglossal duct
from the foramen cecum to the usual location of the
gland. However, with lingual thyroid tissue, there is
Pearls, Pitfalls, and Frequently Asked Questions 545
usually a failure of normal development and no tis-
sue in the normal location of the thyroid.
Q: What is meant by the “organification”of iodine?
A: In thyroid metabolism, iodide is oxidized to iodine
and incorporated into tyrosine to form either
monoiodotyrosine or diiodotyrosine.A deficiency in
peroxidase, which catalyzes the reaction, is a cause
of congenital hypothyroidism.
Q: What is the difference in mechanism of thyroid
uptake between Tc-99m pertechnetate and radio-
iodine?
A: Radioiodine is taken up or extracted (trapped) by
the thyroid follicular cell and organified, binding to
tyrosine residues on thyroglobulin and stored in col-
loid of the follicle.Tc-99m pertechnetate is trapped
but not organified.
Q: What has happened to the range for normal percent
thyroid uptake of radioiodine in the United States
over the last 50 years?
A: The normal range has dropped significantly owing to
iodination of salt and the use of iodine in other foods.
In many laboratories,the range was 20–45% as recently
as the mid-1960s but is now 10–30% at 24 hours.
Pearl
Radioiodine is administered orally.Tc-99m pertechnetate
is administered intravenously.
Pitfall
A potentially serious pitfall is to confuse microcuries
with millicuries.
Pearl
The following are the approximate adult doses of I-131
and I-123 used for uptakes, scans, and therapy. Serious
consequences can result from confusing these doses,par-
ticularly if a therapeutic dose is administered instead
of a diagnostic dose. I-131 uptake (10 microcurie [μCi]),
I-123 uptake (100 μCi), I-123 scan (400 μCi), I-131 scan
(50 μCi), I-131 therapy for Graves’ disease (10 millicuries
[mCi]), and thyroid cancer (75–200 mCi).
Q: What is the normal distribution of radioiodine and
Tc-99m pertechnetate?
A: Radioiodine is taken up by the thyroid, salivary
glands, stomach, and excreted by the kidneys.Tc-99m
pertechnetate has identical uptake and clearance,
except that it is not organified, and thus remains in
the thyroid for a considerably shorter time.
546 NUCLEAR MEDICINE: THE REQUISITES
Q: Which are common causes of falsely low thyroid
uptakes?
A: Patients taking thyroid hormones, iodine-containing
drugs, or recent administration of intravenous iodine
containing radiographic contrast.
Pearl
Synthroid should be discontinued for 4 weeks prior to
a thyroid uptake or scan and Cytomel 2 weeks prior. CT
intravenous iodine contrast agents should not have been
received within 6–8 weeks.
Q: Which drugs are used clinically to block unwanted
thyroid uptake of radioiodine, such as from admin-
istered diagnostic I-131 MIBG (metaiodobenzylguani-
dine) or therapeutic I-131 Bexxar (tositumomab)?
A: Iodine as supersaturate potassium iodide (SSKI) or
perchlorate, a nonvalent ion, that competitively
binds iodine trapping.
Q: What is the difference between and thyroid scan
and thyroid uptake?
A: A thyroid uptake is usually a nonimaging study using
a gamma-detector probe, whereas a thyroid scan
results from gamma-camera imaging.
Pearl
Swallowed activity from salivary secretions on radio-
pertechnetate scans occasionally remains in the esopha-
gus and can be confusing.The nature of the activity is
readily established by having the patient drink water, fol-
lowed by reimaging of the thyroid gland.
Q: How can a thyroid uptake test differentiate the two
most common causes of thyrotoxicosis, Graves’ dis-
ease and subacute thyroiditis? Why?
A: In the initial phase of subacute thyroiditis, thyroid
hormones are released from the inflamed gland
causing thyrotoxicosis. Due to pituitary feedback,
TSH is suppressed. Radioiodine or Tc-99m uptake
requires TSH stimulation. Thus, the uptake of
radioiodine or Tc-99m pertechnetate is low or sup-
pressed. With Graves’ disease, TSH is suppressed;
however, the gland is autonomous and the uptake is
high.
Q: What is the mechanism of action of antithyroid
drugs propylthiouracil (PTU) and methimazole
(Tapazole)?
A: Both PTU and methimazole are thiourea antithyroid
drugs that block the organification of iodine.
Q: What medical conditions are associated with an
increased incidence of paragangliomas (pheochro-
mocytomas)?
A: Both forms of multiple endocrine neoplasia type II
are associated with pheochromocytoma, as are von
Hippel-Lindau disease and neurofibromatosis.
Pitfall
Autonomous nodules are not synonymous with toxic nod-
ules.Patients with small autonomous nodules (less than 3 cm
in diameter) are most often euthyroid.The incidence of thy-
roid cancer in a patient with a single cold nodule is 15–20%,
a multinodular goiter,5%,and a hot nodule,less than 1%.
Q: Which radiopharmaceutical is used most commonly
to localize a clinically diagnosed parathyroid ade-
noma? Describe its characteristic and diagnostic
pharmacokinetics.
A: Tc-99m sestamibi is taken up by both thyroid and
hyperfunctioning parathyroid tissue;however,it is typ-
ically cleared faster by the thyroid, thus the rationale
for early (15 minutes) and delayed (2 hour) imaging.
At early imaging uptake in the thyroid is dominant,
whereas a hyperfunctioning parathyroid may not be
apparent, or may be seen as focal hot uptake, particu-
larly if adjacent to the thyroid. On delayed imaging,
only the parathyroid uptake is dominant.
Pearl
The most common false positive for parathyroid scan-
ning is a thyroid adenoma. Benign and malignant tumors
are other causes for false-positive scintigraphy.
BONE
Q: What are the potential impurities in technetium-
labeled diphosphonate compounds, based on their
biodistribution?
A: Activity in the oropharynx, thyroid gland, and stom-
ach suggests free unlabeled Tc-99m pertechnetate.
Activity in the liver suggests a colloidal impurity.
Rarely, activity is seen in the gut, the result of excre-
tion of activity through the biliary system.The mech-
anism is not understood. Other increased soft tissue
or renal activity is usually caused by a disease
process rather than tracer impurity.
Q: What percentage of the Tc-99m-labeled compounds is
retained in the skeleton at the usual time of imaging?
A: In normal adult subjects, 40–60% of the injected
dose is in the skeleton 2–3 hours after tracer admin-
istration.

Pitfall
The greatest pitfall in interpreting bone scans is failure to
understand its inherent nonspecificity. In our zeal “not to
miss the cancer,” many incidental areas of abnormally
increased tracer accumulation are incorrectly attributed
to metastatic disease.The most common pitfalls are diag-
nosing areas of arthritis or prior trauma as metastases.
Correlative radiographs are often indicated.
Q: Which factors favor osteoarthritis versus metastatic
disease as the cause of increased activity?
A: Osteoarthritis has characteristic locations in the
extremities. Because metastatic lesions are relatively
rare below the proximal femurs or beyond the prox-
imal humeri, osteoarthritis should be considered
first in the elbows, wrists, hands, knees, and feet of
older patients. Involvement of both sides of a joint is
common in arthritis but unusual in metastatic dis-
ease.The lower lumbar spine is the most problem-
atic area because both arthritis and metastases are
common there.
Q: What is the distribution of metastatic deposits from
epithelial primary malignancies in the skeleton?
A: A rule of thumb is that 80% of metastases are found in
the axial skeleton (spine, pelvis, ribs, and sternum).
The remaining metastases are distributed equally
between the skull (10%) and the long bones (10%).
Pearl
Metastases from lung cancer are the most common cause
of metastases in the distal extremities, such as the hands.
Pearl
The majority of epithelial tumor metastases localize first
in the red marrow.The skeletal tracers do not localize in
the tumor tissue but rather in the reactive bone around
the metastatic deposits.
Pitfall
A small amount of activity is frequently seen at the injec-
tion site; this should not be confused with a metastatic
lesion.Likewise,variable degrees of urinary contamination
on the skin may be superimposed on skeletal structures
and confused with activity caused by metastatic disease.
Pearl
In many diseases, the bone scan has a very high sensitiv-
ity for detection of bone metastases. Sensitivity is lower
in tumors with a lytic rather than blastic response, such
Pearls, Pitfalls, and Frequently Asked Questions 547
as multiple myeloma, thyroid cancer, renal cell carci-
noma.The bone scan is also less sensitive for tumors that
preferentially go to bone marrow, such as lymphoma.
Q: How can the radiation dose to the bladder, ovaries,
and testes be reduced?
A: The radiation dose to these structures is largely
caused by radioactivity in the bladder.Frequent void-
ing reduces the radiation dose.
Q: What factors distinguish a superscan resulting from
metastatic disease from a superscan resulting from
metabolic disease?
A: In the usual superscan resulting from metastatic dis-
ease, the increased uptake is restricted to the axial
skeleton and the proximal parts of the femurs and
humeri, the red marrow-bearing areas. In metabolic
bone disease, the entire skeleton is typically affected
with increased uptake seen in the extremities as well
as in the axial skeleton. In some cases resulting from
secondary hyperparathyroidism, increased activity
will also be seen in the lung and stomach.
Pearl
Faint or absent visualization of the kidneys is one of the
findings on superscans that should alert the observer.
This may be misinterpreted as indicating lack of excre-
tion of tracer through the kidneys. In cases of superscan
resulting from metastatic disease, visualization of the kid-
neys is faint because: (1) the skeleton accumulates more
tracer than usual, leaving less available for renal excre-
tion, and (2) owing to the increased skeletal tracer
uptake, the renal activity may actually fall below the gray-
scale threshold.The presence of renal activity is readily
established by adjusting the intensity setting window.
Q: What is the mechanism of the “flare”phenomenon?
A: In some patients treated with chemotherapy for
metastatic disease, regression of the tumor burden is
associated with increased osteoblastic activity, pre-
sumably caused by skeletal healing in response to
chemotherapy. This can appear on skeletal scinti-
grams as a paradoxical increase or apparent “wors-
ening”of the abnormal tracer uptake,which may last
for up to 6 months after therapy.
Q: What is the postmastectomy appearance of the thorax?
A: With radical mastectomy, the majority of the soft
tissue is removed from the corresponding anterior
thorax.The ribs appear “hotter”than on the contralat-
eral side.This is probably caused by less attenuation
of rib activity by soft tissue. Note, however, that if the
patient is imaged with a prosthesis in place, the rib
activity may be attenuated.
548 NUCLEAR MEDICINE: THE REQUISITES
Q: What factors contribute to prolonged fracture posi-
tivity on scintigrams?
A: Displaced and comminuted fractures and fractures
involving joints tend to have prolonged positivity
scintigraphically. Elderly patients have delayed healing.
Q: What factors favor shin splints versus stress fracture
scintigraphically in the tibia?
A: Stress fractures are classically focal or fusiform.The
uptake can involve the entire width of the bone or
extend partially across the shaft of the bone. Shin
splints are classically located along the posterior
medial tibial cortex and involve a third or more of
the length of the bone. In pure shin splints, a focal
component should not be present and superficial lin-
ear activity runs parallel to the long axis of the bone.
Q: The three-phase bone scan is used to diagnose
osteomyelitis.What are other causes for a positive
three-phase scan?
A: Recent fracture, tumor, Charcot’s joint, and soft-
tissue infection overlying chronic noninfectious
bone disease.
Pitfall
False-negative scintigrams may be seen in neonates with
osteomyelitis. Neonates may even have cold lesions. False
negatives may also be seen in very old or debilitated
patients and in patients who have received a course of
antibiotic therapy before scintigraphy is performed.
HEPATOBILIARY
Q: What are the two FDA-approved technetium-99m
iminodiacetic acid analog (IDA) radiopharmaceuti-
cals in clinical use, and how are they different?
A: Tc-99m DISIDA (disofenin) and Tc-99m mebrofenin
(Choletec).The latter has better hepatic extraction,
98% versus 88%, and less renal excretion, 1% versus
9%.The higher extraction of mebrofenin is prefer-
able in patients with hepatic insufficiency.
Pearl
Tc-99m IDA is extracted by the same cellular mechanism
as bilirubin but it is not conjugated. The radiopharma-
ceutical then follows the path of bile through the biliary
system into the bowel.
Pearl
The alternative route of excretion for Tc-99m IDA radio-
pharmaceuticals is via the kidneys.The amount of excre-
tion is usually small but increases with hepatic dysfunction.
Q: What is the most important question to ask a patient
before starting cholescintigraphy for suspected
acute cholecystitis, and why?
A: “When did you last eat?” If the patient has eaten in
the last 4 hours, the gallbladder may be contracted
secondary to endogenous stimulation of cholecys-
tokinin (CCK), and therefore radiotracer cannot
gain entry into the gallbladder. If the patient has not
eaten in more than 24 hours, the gallbladder may
not have had the stimulus to contract and will be
full of thick, concentrated bile, which may prevent
tracer entry.
Pearl
It is also important to ask what the patient ate.The meal
must have contained 10 grams of fat in order to contract
the gallbladder.
Q: What are four indications for CCK infusion?
A: (1) To empty gallbladder in a patient fasting longer
than 24 hours. (2) To differentiate common duct
obstruction from functional causes. Delayed imag-
ing could be used as an alternative. (3) To exclude
acute acalculous cholecystitis if the gallbladder fills
in a patient strongly suspected of having the disease.
A diseased gallbladder will not contract, due to
either acute or chronic disease. (4) To confirm or
exclude chronic acalculous cholecystitis.
Q: Cholescintigraphy is a very sensitive and specific
test for acute cholecystitis. In what clinical settings
is there an increased incidence of false positive cho-
lescintigraphy for acute cholecystitis?
A: In patients who have fasted less than 4 hours or
more than 24 hours, those receiving hyperali-
mentation, and those who have chronic cholecysti-
tis, hepatic dysfunction, or concurrent serious
illness.
Q: What is the rim sign seen with cholescintigraphy,
and what is its significance?
A: The rim sign is increased activity in the liver adja-
cent to the gallbladder fossa.This finding has been
associated with an increased incidence of the com-
plications, such as perforation and gangrene.
Pearl
Increased blood flow to the region of the gallbladder as
a result of severe inflammation is sometimes seen with
acute cholecystitis. In these cases, the rim sign is usually
also seen.

Q: At what time after Tc-99m IDA injection is nonfilling
of the gallbladder diagnostic of acute cholecystitis?
A: One hour is defined as abnormal. However, nonfill-
ing of the gallbladder is diagnostic of acute cholecys-
titis if delayed images show no filling by 3 to 4 hours
or 30 minutes after morphine administration.
Pearl
Delayed visualization is most commonly seen in chronic
cholecystitis. It is also seen with hepatic dysfunction
caused by altered pharmacokinetics, which is delayed
uptake and clearance.
Q: What is the mechanism of morphine-augmented
cholescintigraphy?
A: Morphine increases tone at the sphincter of Oddi,
producing increased intraductal pressure. This
results in bile flow preferentially through the cystic
duct, if it is patent.
Q: What is the most common cholescintigraphic find-
ing in chronic cholecystitis during routine cho-
lescintigraphy?
A: A normal study.Less than 5% of patients with chronic
cholecystitis have delayed filling. Other less common
associated findings include delayed biliary-to-bowel
transit time and, rarely, nonvisualization of the gall-
bladder or intraluminal filling defects. A reduced
gallbladder ejection fraction is seen with sympto-
matic chromic cholecystitis.
Q: What is acute acalculous cholecystitis?
A: Acute cholecystitis without a stone occluding the
cystic duct.The obstruction may be caused by debris
or inflammatory changes, or the cholecystitis may be
limited to the gallbladder wall because of infection,
ischemia, or toxemia.This disease occurs in very sick
hospitalized patients who have sustained trauma,
burns, sepsis, or other serious illness and is associ-
ated with a high morbidity and mortality.
Pearl
The sensitivity of cholescintigraphy for acute acalculous
cholecystitis is probably only approximately 75–85%
compared with 98% for acute calculous cholecystitis.
Pearl
If the clinical suspicion for acute acalculous cholecystitis is
high but the gallbladder visualizes, sincalide can be helpful
diagnostically. Cholecystitis can be excluded if the gallblad-
der contracts. If it does not contract, the cause could be
acute or chronic acalculous cholecystitis.A radiolabeled
leukocyte study can confirm acute disease.
Pearls, Pitfalls, and Frequently Asked Questions 549
Q: The diagnosis of common duct obstruction is com-
monly made by sonographic detection of a dilated
common duct. In what clinical situations would cho-
lescintigraphy be useful?
A: In early acute obstruction before the duct has had
time to dilate (24–48 hours), and in patients with
previous obstruction or ductal instrumentation who
have baseline dilated ducts. In both these situations,
cholescintigraphy can be diagnostic.
Q: What are cholescintigraphic findings of high-grade
common duct obstruction?
A: Prompt hepatic uptake but a persistent hepatogram
without clearance into biliary ducts because of the
high backpressure.
Q: What are the cholescintigraphic findings of partial
common duct obstruction?
A: Retention of activity in the biliary ducts, delayed
biliary-to-bowel clearance, and poor ductal clear-
ance on delayed imaging or with sincalide.
Pearl
Delayed biliary-to-bowel transit is an insensitive and
nonspecific finding for common duct obstruction.
Delayed biliary to bowel clearance is seen in only 50%
of patients with obstruction. On the other hand,
delayed biliary-to-bowel transit may be seen in 20% of
healthy subjects. It is also seen in patients pretreated
with sincalide.
Pearl
Administration of sincalide will cause sphincter of Oddi
relaxation, prompt biliary duct clearance and biliary-
tobowel transit in patients with functional causes, but
will remain abnormal in patients with partial common
duct obstruction.
Pitfall
The methodology use for administering sincalide is diag-
nostically important. A bolus infusion may cause spasm
of the neck of the gallbladder and ineffective emptying.
Similarly, 1–3 minute infusions may result in poor con-
traction of the gallbladder in approximately one-third of
normal subjects. Sincalide, 0.01–0.02 μg, should be
infused slowly over 30–60 minutes.
Q: What ancillary maneuver increases the sensitivity of
cholescintigraphy for detection of biliary atresia?
A: The administration of phenobarbital for 3–5 days
before the HIDA activates the liver enzymes.
A serum phenobarbital level should be in the thera-
peutic range before cholescintigraphy is started.
550 NUCLEAR MEDICINE: THE REQUISITES
Q: How is the diagnosis of biliary atresia made with
cholescintigraphy?
A: No clearance of Tc-99m IDA tracer is seen by
24 hours. Biliary clearance is consistent with other
etiologies for neonatal hepatitis.
Q: What is the postcholecystectomy syndrome and
what are common causes for it?
A: Recurrent biliary colic-like pain after cholecystec-
tomy. Cystic duct remnant, retained or recurrent
stone, inflammatory stricture, sphincter of Oddi dys-
function.
Pearl
Sphincter of Oddi dysfunction is essentially a partial biliary
obstruction at the level of the sphincter without evidence
of stone or stricture.Cholescintigraphy will show a pattern
of partial biliary obstruction.The diagnosis is ultimately
made by excluding stones or stricture with ERCP and the
finding of elevated sphincter pressure with manometry.
Q: What is the difference in clinical presentation and
clinical course of patients with focal nodular hyper-
plasia (FNH) and hepatic adenoma?
A: FNH is asymptomatic and found incidentally,
whereas hepatic adenomas often present with hem-
orrhage can be life-threatening. Adenomas are
closely associated with the use of oral contracep-
tives and they must be discontinued.
Q: What are the Tc-99m sulfur colloid scintigraphic
findings in FNH and hepatic adenoma?
A: Hepatic adenomas do not usually show Tc-99m
sulfur colloid uptake because they typically
do not have Kupffer cells. FNH is associated with
increased blood flow.With FNH, uptake is normal or
increased in two thirds of patients; however, one
third have no uptake.
Q: What are the cholescintigraphic findings in FNH,
hepatic adenoma, and hepatoma?
A: FNH shows increased flow, normal uptake, and
delayed focal clearance. Hepatomas are cold on early
images but often fill on delayed images (2 hours).
The hepatoma is functional, but hypofunctional
compared with the normal liver. Hepatic adenomas
do not typically have uptake.
Q: The specificity of Tc-99m-labeled red blood cells for
diagnosis of cavernous hemangioma is very high.
False positives are rare.What factors affect the sensi-
tivity of the test?
A: Small size (less than 1.5 cm), attenuation (small cen-
tral hemangiomas are harder to detect than superfi-
cial ones, hemangiomas adjacent to major vessels
may be harder to detect), and methodology (SPECT
is more sensitive than planar imaging).
Q: Which of the following statements is true in regard
to the diagnosis of hemangiomas?
a. Ultrasonography is neither sensitive nor specific.
b. CT is not very sensitive when strict criteria
are used and not specific when liberal criteria
are used.
c. MRI is sensitive, has a distinctive pattern (light
bulb sign on T-2 weighted images), and is much
more specific than CT or ultrasonography,
although various other benign and malignant
tumors may have an appearance similar to
hemangioma.
d. MRI is the method of choice for small lesions
adjacent to large vessels.
A: All are true.
Q: What are the characteristic scintigraphic findings in
liver hemangioma?
A: Blood flow is normal.Immediate images show a cold
defect, whereas delayed images acquired 1–2 hours
after tracer administration show increased uptake
within the lesion compared with the normal liver,
often equal to uptake in the spleen and heart. SPECT
is mandatory for smaller lesions.
Q: In regard to regional intraarterial chemotherapy,
which of the following statements is/are true?
a. Hepatic arterial chemotherapy preferentially
perfuses the tumor, with relative sparing of unin-
volved liver.
b. Systemic toxicity is directly related to the
amount of chemotherapeutic agent that reaches
the systemic circulation.
c. The response to therapy can be predicted from
Tc-99m macroaggregated albumin (MAA) hepatic
arterial perfusion scintigraphy.
d. Symptoms of drug toxicity can be easily differen-
tiated clinically from the progression of liver
metastases.
A: a. True.Tumor in the liver receives its blood supply
primarily from the hepatic artery; the normal
liver receives approximately 70% of its blood
supply from the portal vein.
b. True. For example, arteriovenous shunting will
result in systemic exposure and toxicity.
c. True. Evidence of proper catheter placement
and perfusion of tumor nodules is associated
with a good response to therapy.

d. False. Symptoms are identical. Only the Tc-99m
MAA study can make that differentiation by
determining the adequacy of perfusion and the
presence or absence of extrahepatic perfusion.
Q: What is the significance of the extrahepatic perfu-
sion seen on Tc-99m MAA hepatic arterial per-
fusion studies in patients receiving intra-arterial
chemotherapy for liver metastases?
A: Extrahepatic perfusion of abdominal viscera, most
often the stomach but also the bowel, pancreas, and
spleen,is associated with a high incidence of adverse
symptoms (nausea, vomiting, abdominal pain), about
45%, versus a 16% incidence of similar symptoms in
patients treated identically but without evidence of
extrahepatic perfusion on the Tc-99m MAA study.
GENITOURINARY
Q: What percentage of renal plasma flow is filtered
through the glomerulus, and what percentage is
secreted by the tubules?
A: Twenty percent of renal plasma flow is cleared by
glomerular filtration and 80% by tubular secretion.
Q: Which nonradioactive drugs used to calculate
glomerular filtration rate (GFR) and effective renal
plasma flow (ERPF) are considered to be the refer-
ence standards?
A: Inulin for GFR and paraaminohippurate (PAH) for
ERPF.
Q: Which radiopharmaceuticals are most often used
clinically for measurement of GFR and ERPF?
A: Tc-99m diethylenetriamine pentaacetic acid (DTPA)
for GFR. Iodine-131 orthoiodohippurate (OIH) was
used for ERPF in the past.As I-131 OIH is not avail-
able commercially,Tc-99m MAG3 is currently used.
However,Tc-99m MAG3 does not actually measure
ERPF and a correction factor based on proportional
clearance compared with I-131 OIH must be applied.
Some sites report these values not as ERPF but as
MAG3 clearance.
Q: What is the mechanism of renal uptake for I-131
OIH, Tc-99m mercaptylacetyltriglycine (MAG3),
Tc-99m DTPA, Tc-99m dimercaptosuccinic acid
(DMSA), and Tc-99m glucoheptonate (GH)?
A: Tc-99m DTPA, glomerular filtration;Tc-99m MAG3,
tubular secretion; I-131 OIH, tubular secretion and
glomerular filtration; Tc-99m GH, cortical binding
and glomerular filtration; and Tc-99m DMSA, cortical
binding.
Pearls, Pitfalls, and Frequently Asked Questions 551
Q: What is the percent cortical binding of Tc-99m
DMSA and Tc-99m GH?
A: Tc-99m DMSA, 40–50%;Tc-99m GH, 10–20%.
Pearl
The two radiopharmaceuticals bind to the proximal con-
voluted tubules in the cortex.
Q: What is Webster’s rule?
A: Pediatric radiopharmaceutical doses can be estimated
using the formula (age + 1)/(age + 7) × adult dose.
Q: The time-to-peak activity of a renal time-activity
curve (TAC) represents which of the following:
a. The end of extraction
b. The beginning of renal clearance
c. The time point at which the amount of cortical
uptake of the radiopharmaceutical is equal to
clearance
A: c. Uptake and clearance are occurring simultane-
ously over a period because of several factors
that include an imperfect bolus, the percent
first-pass extraction fraction of the radiotracer,
the amount of recirculating radiotracer, and the
normal variability of nephron function.
Pearl
Dehydration does not affect the first part of the time
activity curve while later phases may be altered by many
factors including dehydration and hydronephrosis.
Q: What is the proper renal region of interest (ROI)
selection on the computer for the following:
a. Diuresis renography
b. Captopril renography
A: a. The ROI should include the dilated pelvis and
the cortex. Because of hydronephrosis, the
dilated collecting system counts predominate.
b. A whole kidney ROI is adequate if there is no
pelvic retention. Lasix is often given with the
radiopharmaceutical to ensure pelvicocalyceal
clearance.When there is pelvicocalyceal activity,
a peripheral two-pixel cortical ROI should be
selected to avoid the effect of these counts on
the TAC.A drop in GFR with captopril is mani-
fested as deterioration in the cortical TAC
(delayed peak and decreased function) when
using Tc-99m DTPA or cortical retention when
the radiotracer is Tc-99m MAG3.An identical ROI
should be used for the baseline comparison
study.
552 NUCLEAR MEDICINE: THE REQUISITES
Q: Differential renal function is evaluated by drawing
kidney and background ROIs.The relative uptake of
the two kidneys after background correction is deter-
mined.Which time interval is used to calculate differ-
ential renal function for dynamic renal scintigraphy?
a. Entire 30-minute study
b. The 60-second flow study
c. The time after the initial flow study
d. Interval of 1–3 minutes
A: d. Because cortical uptake of the renal radiophar-
maceutical is of interest, the optimal interval is
after the initial flow but before the collecting
system activity appears, usually 1–3 minutes.
With good function, activity may be seen
before 3 minute, especially in children. Radio-
pharmaceuticals with higher extraction also
clear faster.Thus the 1–2 minute interval may be
optimal overall. Ideally the clinician should
review the dynamic frames to determine when
calyceal clearance occurred and use the 60- to
90-second interval before that.
Q: What are the general methods for calculating
absolute GFR?
A: Blood sampling, blood sampling and urine collec-
tion, and camera-based methods.
Q: At what step in the renin-angiotensin-aldosterone
cascade does captopril work? In which organ does
this occur?
A: Captopril blocks the conversion of angiotensin I to
angiotensin II in the lungs.
Pearl
The usual captopril dose, 25 mg, although pharmaco-
logically effective on the renal vasculature, is usually
inadequate to produce peripheral vasodilation and hypo-
tension. However, a patient may rarely develop hypoten-
sion, requiring prompt fluid administration to maintain
intravascular volume and pressure.
Q: In renal artery stenosis the effect of captopril is mani-
fested by a reduction in blood flow to the kidney that
can be seen on radionuclide angiography. True or
false?
A: False. Blood flow is not affected by captopril. If it is
poor to begin with, it will remain poor. If it is nor-
mal, no change is seen.The compensatory mecha-
nism for maintaining the glomerular filtration rate is
renin dependent and results in decreased GFR after
captopril administration.
Q: Which of these factors affects the accuracy of diure-
sis renography?
a. State of hydration
b. Renal function
c. Dose of diuretic
d. Radiopharmaceutical
e. Bladder capacity
A: All of the above.Adequate hydration is required for
good urine flow and adequate response to the
diuretic. A full bladder may cause a functional
obstruction. Intravenous hydration and urinary
catheterization are strongly suggested, especially in
children. Tc-99m DTPA, Tc-99m MAG3, and I-131
OIH have all been successfully used. Because of its
better extraction efficiency and good image resolu-
tion,Tc-99m MAG3 is the agent of choice in renal
insufficiency.Tc-99m DTPA works well in patients
with good renal function. Renal insufficiency is
a definite limitation to diuresis renography.The kid-
ney must be able to respond to the diuretic chal-
lenge. Therefore the dose of diuretic must be
increased in renal insufficiency, but the exact dose
required is only an educated estimate.
Q: A good diuretic response rules out a partial obstruc-
tion.True or false?
A: False. Diuretic renography is often performed to
determine the functional significance of a known
partial obstruction, such as in patients with cervical
or bladder cancer.A poor diuretic response indicates
that intervention is indicated.With good clearance,
no immediate intervention is required. Follow-up
evaluations may be indicated.
Q: What is the most sensitive technique for diagnosing
scarring secondary to reflux?
A: Tc-99m DMSA cortical imaging. Ultrasonography
and intravenous urography have considerably lower
sensitivity.
Q: How can radionuclide imaging differentiate upper
from lower urinary tract infection, and why is this
differentiation important?
A: With upper tract infection or pyelonephritis,Tc-99m
DMSA shows tubular dysfunction, manifested by
decreased uptake.This is a reversible process.With
appropriate therapy, tubular will return in 3–6
months. Upper tract infection has prognostic impli-
cations because it may lead to subsequent renal
scarring, hypertension, and renal failure.

Q: Why is radionuclide cystography preferable to the
contrast method in most cases? What is the exception?
A: The radionuclide test is more sensitive for detec-
tion of reflux than contrast-enhanced voiding cys-
tourethrography and results in much less radiation
exposure (50- to 200-fold less) to the patient.The
only exception is in the first evaluation of a male,
when the better resolution of the contrast study can
permit the diagnosis of an anatomical abnormality
such as posterior urethral valves.
Pearl
Scintigraphy allows for calculation of bladder volumes
and residuals.The residual activity in the bladder is calcu-
lated by one of two methods using a region of interest
(ROI) around the bladder:
Residual volume (ml) = [voided urine volume (ml) ×
postvoid bladder counts]/
[initial bladder counts –
postvoid bladder counts]
or
Residual bladder volume (ml) = [postvoid bladder counts ×
volume infused]/
initial bladder counts
Q: What is meant by “direct” versus “indirect” radionu-
clide cystography and which is the preferred
method for detecting vesicoureteral reflux?
A: Direct cystography, that is, cystography requiring uri-
nary tract catheterization and infusion of radiotracer
into the bladder,is a more sensitive method for detect-
ing vesicoureteral reflux. Reflux can be detected dur-
ing bladder filling as well as voiding. In contrast, the
indirect method, where a routine renogram is initially
performed,cannot be used to detect reflux during the
bladder filling stage because radiotracer is flowing
through the collecting system antegrade.
Q: What is the most common developmental abnormal-
ity leading to testicular torsion?
A: The bell-clapper testis.
Pearl
The bell-clapper testis is a congenital abnormality and
usually bilateral. Prophylactic surgery is performed on
the asymptomatic side.
Q: What is the difference in blood supply to the testes
and scrotum?
A: The testes receive blood predominantly from the
testicular artery, whereas the scrotum receives its
supply from the pudendal vessels.
Pearls, Pitfalls, and Frequently Asked Questions 553
ONCOLOGY—POSITRON
RADIOPHARMACEUTICALS
Q: Which of these statements is true regarding fluo-
rine-18 fluorodeoxyglucose (FDG) PET?
a. F-18 FDG uptake is normally high in the brain
and heart.
b. The mechanism of F-18 FDG uptake and metabo-
lism is identical to that of glucose.
c. Oncology patients should fast for at least 4 hours
prior to injection.
d. Unlike glucose, F-18 FDG is excreted through
the genitourinary system.
A: a, c, d.F-18 FDG enters the cell in a fashion similar to
glucose but becomes trapped within the cell
because it cannot progress further through the glu-
cose enzymatic pathways.
Q: The most common indication for F-18 FDG PET is
for the staging of lung cancer.Which of the follow-
ing statements is false?
a. The sensitivity for detection is high for tumors
of 5 mm and greater in size.
b. Sensitivity and specificity of FDG PET is higher
than CT for mediastinal staging of lung cancer.
c. False negatives may be seen with hyperglycemia.
d. False negatives may occur with bronchoalveolar
carcinoma.
A: a. The sensitivity for detection is reduced for
tumor size less than 8–10 mm.
Q: The sensitivity of FDG PET is high for detection of
many malignancies. For which of the following
tumors is the sensitivity of FDG PET not high?
a. Colorectal cancer
b. Melanoma
c. Hepatocellular carcinoma
d. Renal cell carcinoma
e. Lymphoma
A: FDG PET has poor sensitivity for primary hepatocel-
lular carcinoma, renal carcinoma, as well as prostate
cancer.This is less true of metastatic disease than pri-
mary tumors.
Pearl
For thyroid cancer imaging, I-131 is more sensitive than
F-18 FDG for well-differentiated papillary or follicular thy-
roid carcinoma. In patients who have been treated with
I-131 and have a negative I-131 whole-body scan on
follow-up evaluation but have an elevated serum thy-
roglobulin, F-18 FDG PET has good sensitivity for detec-
tion of malignancy.The reason is that in this setting, the
tumor has dedifferentiated into a higher grade malignancy.
554 NUCLEAR MEDICINE: THE REQUISITES

Q: What are the advantages of CT PET over PET alone?

a. Automated hardware and software fusion for Pearl
anatomical localization
PET can help direct biopsy to the most metabolically
b. Diagnostic CT at the same time as the PET scan
active area of a mass to help avoid sampling areas of
c. CT is used for more rapid attenuation correction
necrosis.
d. Eliminates false positive interpretations
A: a, c. Diagnostic CT (b) often requires contrast.
Q: Which is the best modality for the detection of
Furthermore, to minimize radiation dose to the
osseous metastases?
patient, 80 mA is commonly used, which reduces
A: It depends. MRI is highly sensitive and often
diagnostic quality. CT PET reduces false positives,
detects lesions not seen on bone scan as it can visu-
but does not eliminate them (d).
alize changes in the marrow and does not depend
on secondary reactive cortical bone changes to
Pearl develop. Bone scan, on the other hand, can image
the whole body in a cost-effective manner. It is very
With CT PET,misregistration due to patient motion,respi-
useful in tumors with sclerotic metastases. PET
ratory motion, organ movement (bowel) can introduce
scanning with F-18 FDG is more sensitive than
potential false positive interpretations. PET is acquired at
bone scan for more aggressive lytic tumors. At
normal tidal volume breathing. CT with PET may be
times the two modalities complement each other
acquired with breath-hold or shallow breathing, leading
by detecting different lesions in the same patient.
to artifacts due to errors in anatomical registration and
F-18 sodium fluoride PET is a very sensitive bone
attenuation correction.
scan agent and some feel that it may replace Tc-99m
MDP imaging. Each of these methods is more sensi-
Q: What are some limitations of FDG PET in tumor stag-
tive than CT or radiographs.
ing?
A: PET imaging does not detect microscopic metas-
Q: What are some differences between nonattenuation-
tases, tumor involvement in local lymph nodes may
corrected PET images and attenuation-corrected
be obscured by activity in an adjacent tumor, con-
images?
current infection/inflammatory processes may cause
A: The noncorrected image has a very different
false positives, and sensitivity for intracranial metas-
appearance than the corrected image. Structures
tases is low.
near the surface appear more intense as fewer pho-
tons are attenuated before hitting the detector.This
Q: What are limitations of tumor restaging by PET?
explains why the skin looks like it is outlined with
A: Posttherapy effects of surgery, chemotherapy, and
a charcoal pencil. The air-filled lungs are also
especially radiation therapy may cause increased
intense. Because fewer counts are seen in central
F-18 FDG uptake, which can be confused with activ-
areas, lesions may be missed. For accurate quantifi-
ity from active tumor. Even patients scheduled for
cation (SUV), the attenuation corrected images
imaging after an appropriate delay after therapy may
must be used.
require follow-up imaging. If activity is diminishing,
this helps confirm a benign process.
Q: What is the most significant type of scatter experi-
enced in PET?
Pearl A: Compton scatter is most common in the energy
range of PET (511 keV). It is particularly a problem
The usual recommended FDG PET imaging time to evalu-
with three-dimensional (3-D) mode (no septa) acqui-
ate response therapy after chemotherapy is 3 weeks, but
sition. 3-D is highly sensitive and faster than two-
at least 2–3 months for radiation therapy. It is not always
dimensional, but also accepts more scatter counts
possible from a clinical standpoint to follow these guide-
which causes reduced image quality.
lines, but an awareness of the potential problem is criti-
cal for interpretation.

Q: What is the role of F-18 FDG PET in differentiating

infection from tumor in a patient with AIDS and ONCOLOGY—SINGLE PHOTON
numerous parenchymal lung abnormalities on CT? RADIOPHARMACEUTICALS
A: The use of PET is very limited as it can not differenti-
ate infection from tumor, as both can show intense Q: What is the mechanism of gallium-67 uptake in
radiotracer uptake. tumors?

A: Ga-67 binds to serum iron transport molecules
such as transferrin, which transports the Ga-67 to
the tumor. Ga-67 enters the extracellular fluid
space via the tumor’s leaky capillary endothelium.
It is bound to the tumor cell surface by transferrin
receptors and then transported into the cell, where
it binds to proteins such as ferritin and lactoferrin,
which are in increased concentration in tumors.
Q: Ga-67 uptake is normally seen in which of the fol-
lowing organs: salivary glands, lacrimal glands, thy-
mus, spleen, breast, heart?
A: Salivary gland and lacrimal gland uptake is normal
and variable.Thymus uptake may be seen normally
in children, especially after they have received
chemotherapy.The spleen has uptake, but it is low
level. Breast uptake is variable and is most promi-
nent post-partum. Heart visualization is not normal,
but may occur with myocarditis or pericarditis.
Pitfall
Surgical wounds normally have increased uptake for 1–2
weeks postoperatively, and faint activity may remain for
3–4 weeks. Focal bone uptake may be seen after bone
marrow biopsy. Increased bone uptake may also be seen
wherever there is increased bone turnover, such as frac-
tures, orthopedic hardware, arthritis.
Q: For which malignant diseases has Ga-67 been found
most clinically useful for diagnosis, staging, and
restaging?
A: Hodgkin’s disease, malignant lymphoma, hepatoma,
and melanoma. However, F-18 FDG has to a large
extent replaced Ga-67 for this purpose.
Q: Which of the following statements is associated
with Hodgkin’s disease and which with non-
Hodgkin’s lymphoma?
a. Orderly contiguous spread of lymph node
involvement in young patients
b. Multicentric disease with a highly variable clini-
cal course and a high incidence of extranodal
tumor involvement
c. Mediastinal masses are common
d. Abdominal involvement of mesenteric and
retroperitoneal nodes is common
e. High cure rate
f. Variable clinical course that can be indolent or
rapidly lethal
A: Hodgkin’s disease: a, c, e; non-Hodgkin’s lymphoma:
b, d, f.
Pearls, Pitfalls, and Frequently Asked Questions 555
Pearl
Prior to the availability of FDG PET, Ga-67 was used in
patients with Hodgkin’s disease and malignant lym-
phoma posttherapy masses to differentiate residual
tumor or fibrosis, necrosis, or scarring.
Pitfall
A pretherapy Ga-67 study is important for proper evalua-
tion of the posttherapy Ga-67 study.The pretherapy study
ensures that the tumor site is gallium-avid.
Q: Technetium-99m sestamibi has been used for deter-
mination of malignancy of breast masses detected
with mammography or by palpation.Which of these
statements is true?
a. Its accuracy is higher for palpable than for non-
palpable masses.
b. Its sensitivity is poor for lesions less than 1 cm in
size.
c. Fibroadenomas are always negative.
d. It is particularly useful in patients with dense
breasts or those with architectural distortion,
such as previous surgery, radiation therapy, and
breast implants.
A: a, b, d. Fibroadenomas are a common cause for false
positives.
Q: Which of the following statements are true of In-111
capromab pendetide (ProstaScint)?
a. Murine monoclonal antibody against a prostate-
specific membrane antigen expressed by more
than 95% of prostate adenocarcinomas.
b. Its main indication is for localization of soft tis-
sue metastases after prostatectomy in patients
with a rising PSA and negative bone scan.
c. Elevated human murine antibody (HAMA) titers
are observed in 50% of patients.
d. SPECT is mandatory for the pelvis.
A: a,b,d. HAMA elevations are seen in less than 10% of
patients.
Pearl
The membrane specific antigen that In-111 ProstaScint
localizes to is not PSA but rather prostate specific mem-
brane antigen (PSMA), a glycoprotein expressed by
prostate epithelium, which is not expressed on any other
adenocarcinomas.
556 NUCLEAR MEDICINE: THE REQUISITES
Pearl
Adverse reactions occur in 4% of patients receiving
ProstaScint. They are usually minor. The incidence of
adverse affects with a second injection is only 5%, thus it
can be used diagnostically a second time.
Pearl
Although In-111 ProstaScint is useful for detecting soft-
tissue metastases, it is not particularly sensitive for detect-
ing bone metastases.The bone scan is more sensitive.
Pearl
ProstaScint accumulation in the prostate bed in a patient
treated with radioactive seed placement is nonspecific.
Uptake in the surgical bed following prostatectomy is
highly suspicious for recurrent tumor.
Q: Which of the following are true statements regard-
ing In-111 OctreoScan?
a. It is a somatostatin receptor imaging agent.
b. The sensitivity for all neuroendocrine tumors is
very high.
c. Highest uptake is seen in the spleen and
kidneys.
d. Only neuroendocrine tumors have somatostatin
receptors.
A: a, c. Although its sensitivity for detection of most
neuroendocrine tumors is very high, it has a poorer
sensitivity for insulinomas and medullary carcinoma
of the thyroid.Somatostatin receptors are found on a
variety of nonneuroendocrine tumors, including
astrocytomas, meningiomas, malignant lymphoma,
and breast and lung cancer.
Q: In which melanoma patients is sentinel node lym-
phoscintigraphy indicated and why?
A: Patients with a primary lesion less than 1 mm in
thickness are at low risk of recurrence and have a
good prognosis. Patients with a primary lesion
thickness greater than 4 mm are at high risk for
metastatic adenopathy and distant metastases.
Lymphoscintigraphy is indicated for patients with
primary lesions greater than 1 mm and less than
4 mm thickness.
Q: What information does sentinel node lymphos-
cintigraphy provide in patients with intermediate-
thickness malignant melanoma?
A: Sentinel node lymphoscintigraphy can pinpoint the
sentinel node for the surgeon, which can be local-
ized easily at surgery with a gamma probe. After
immunohistochemical staining of tissue from this
lymph node, the presence of metastases can be
determined. The results will determine which
patients require further nodal bed dissection and
adjuvant chemotherapy.
Q: Which radiopharmaceutical is used for melanoma
lymphoscintigraphy? What is the injection method-
ology?
A: Filtered Tc-99m sulfur colloid is the usual agent
because unfiltered Tc-99m SC does not migrate
well from the site of injection. It is injected intracu-
taneously at four sites around the primary lesion
site.
Q: In what other malignant disease, is sentinel node
lymphoscintigraphy commonly performed and how
is it injected?
A: Breast cancer. The method of injection varies.
At some hospitals, it is injected intratumorally.
However, others inject it subdermally, whereas
some inject the Tc-99m SC in the periareolar
region.The rationale is that all lymphatics drain to
the areolar region before drainage to the axillary
region. Lymphatic drainage to the internal mam-
mary or supraclavicular nodes is occasionally
detected.
Pearl
Many experts recommend using two methods to ensure
optimal lymphatic transit. Massaging the site of injection
vigorously after injection promotes migration of the
dose.
Pearl
In patients with AIDS and an intracerebral mass,Tl-201
can differentiate tumor, usually malignant lymphoma,
from inflammatory causes, most commonly toxoplasmo-
sis.Tl-201 is not usually taken up in inflammation, but is
taken up by malignant tumors.Predictive value is approx-
imately 85%.
Q: What is the purpose of the In-111 Ibritumomab
Tiuxetan (Zevalin) scan?
A: This scan assesses biodistribution of the Zevalin
prior to the administration of the therapeutic Y-90
Ibritumomab Tiuxetan (Zevalin).Abnormalities that
indicate therapy should not be given include exces-
sive uptake in the lung or kidneys or activity in these
organs that increases or does not decrease over
time. It is normal but not mandatory to see uptake in
the tumor.

Pearl
The dose of Y-90 Zevalin is adjusted based on the
patient’s platelet count: 0.4 mCi/kg for patients with
platelets >150,000 and 0.3 mCi/kg for platelets between
149,000 to 100,000. If the platelet count is below
100,000, patients should not be treated.
GASTROINTESTINAL
Q: What is achalasia? What is the role of the esophageal
transit study?
A: Achalasia is characterized by absence of peristal-
sis in the distal two-thirds of the esophagus,
increased lower esophageal sphincter (LES) pres-
sure, and incomplete sphincter relaxation after
swallowing. It is associated with symptoms of dys-
phagia, weight loss, nocturnal regurgitation,
cough, and aspiration.The diagnosis is confirmed
by esophageal manometry. Radionuclide esophageal
transit studies have a high sensitivity for making the
diagnosis; however, they are most useful for evaluat-
ing the effectiveness of therapy, such as esophageal
dilation.
Q: Which of the following statements in regard to
reflux and aspiration studies are true or false:
a. The milk study is a sensitive method for diagnos-
ing gastroesophageal reflux.
b. The milk study is a sensitive method for diagnos-
ing aspiration.
c. Frequent image acquisition improves the sensi-
tivity of the milk study.
d. The “salivagram” is a more sensitive method for
diagnosing aspiration.
A: True: a, c, d
False: b. Aspiration is seen only rarely on delayed
imaging.
Q: What is the functional role of the proximal and dis-
tal stomach?
A: The proximal stomach or fundus is responsible for
liquid emptying and for receptive relaxation to
accommodate a large meal. The distal stomach or
antrum is responsible for the grinding and sieving of
solid food and solid emptying.
Q: Describe the difference in emptying patterns
between solids and liquids.
A: Liquids empty exponentially. Solid emptying is bipha-
sic, with an initial lag phase before linear emptying
begins.The lag phase is due to the time required for
Pearls, Pitfalls, and Frequently Asked Questions 557
food to be broken down into small enough pieces to
allow passage through the pylorus.
Q: Which of these factors will affect the rate of gastric
emptying: meal content, time of day, position (stand-
ing, sitting, lying), stress, exercise?
A: All. The gastric emptying study should be standard-
ized for the meal, time of day, patient position,
methodology of acquisition and processing. Normal
values should be derived from this specific protocol.
Q: Which of the following statements are true regard-
ing the need for variable attenuation correction of
gastric emptying studies?
a. Gastric emptying may be underestimated when
an anterior acquisition alone is obtained.
b. The characteristic pattern of attenuation effect
on solid gastric emptying time–activity curve is
a rise in activity after ingestion of the meal
before emptying begins.
c. The geometric mean method is considered the
standard method for attenuation correction.
A: All. Both anterior and posterior acquisitions are
required to correct using the geometric mean calcu-
lation (square root of the product of the anterior and
posterior views).
Q: What other methods can be used as an alternative
to the geometric mean method of attenuation cor-
rection?
A: Left anterior oblique method. Because the camera
head is positioned roughly parallel to movement of
the stomach contents, from the posterior fundus to
the more anterior antrum, no mathematic correc-
tion is needed.
Q: When might the use of Tc-99m sulfur colloid offer
advantages over Tc-99m red blood cells for the diag-
nosis of acute gastrointestinal bleeding?
A: With very rapid bleeding and vascular instability, the
radiotracer can be injected and the study completed
in 15–20 minutes. No radiolabeling of red cells is
necessary.The patient can then go directly to angiog-
raphy, potentially saving the angiographer time and
contrast.
Q: List in increasing order the labeling efficiency of
methods to label Tc-99m red blood cells: in vivo,
in vitro, and in vivtro.
A: In vivo, 75%; in vivtro or modified in vivo, 85%; and
in vitro, 98%. An in vitro commercial kit method
(Ultratag) for labeling Tc-99m erythrocytes is now
available and is the method of choice, particularly
for gastrointestinal bleeding studies.
558 NUCLEAR MEDICINE: THE REQUISITES
Q: Why is the Tc-99m red blood cell method for detect-
ing gastrointestinal bleeding more sensitive than the
Tc-99m sulfur colloid method?
A: A longer acquisition is possible, usually 90 minutes,
and imaging can be performed for up to 24 hours.
Q: What are the essential criteria needed to confidently
diagnose the site of active bleeding on a radionu-
clide study?
A: A radiotracer “hot spot” appears where there was
none and conforms to bowel activity; the activity
increases over time; and the activity moves ante-
grade or retrograde.
Pitfall
A poor label can result in gastric activity that might be
construed as upper gastrointestinal bleeding or urinary
activity that might be misinterpreted as a bleeding site.
Pearl
Look for thyroid and salivary gland uptake when in doubt
about the presence of free Tc-99m pertechnetate.
Pearl
A lateral view of the pelvis should be routine the end of
the acquisition to differentiate bladder, rectal bleeding,
and penile activity.
Pitfall
Focal activity that does not move may be anatomical (e.g.,
kidney,accessory spleen,hemangioma,varices,aneurysm).
Pearl
Contrast angiography can detect bleeding rates of about
1 ml/min versus 0.1 ml/min for the radionuclide study.
Q: Ectopic gastric mucosa is most often seen clinically
in Meckel’s diverticulum.What other gastric abnor-
malities may contain gastric mucosa?
A: Gastrointestinal duplication,Barrett’s esophagus,and a
retained gastric antrum after gastrectomy. In addition,
ectopic gastric mucosa may occur in gastrogenic cys-
tis found in the pancreas,duodenum,and colon.
Pearl
The mucin cells in the stomach are primarily responsi-
ble for gastric uptake of Tc-99m pertechnetate, not the
parietal cells.
Q: What is the origin of Meckel’s diverticulum?
A: This most common congenital anomaly of the gas-
trointestinal tract results from failure of closure of
the omphalomesenteric duct of the embryo, which
connects the yolk sac to the primitive foregut via the
umbilical cord.
Pearl
This true diverticulum (Meckel’s) arises on the anti-
mesenteric side of the bowel, usually 80–90 cm proximal
to the ileocecal valve, although it can occur elsewhere.
Pearl
Gastric mucosa is present in 10–30% of all Meckel’s diver-
ticula, in 60% of symptomatic patients, and in 98% of
those with bleeding.
Pitfall
A number of false-positive studies have been reported
over the years in scans for Meckel’s diverticula, including
those of urinary tract origin (e.g., horseshoe kidney,
ectopic kidney), those resulting from inflammation (e.g.,
inflammatory bowel disease, neoplasms), bowel obstruc-
tion (seen most often with intussusception and volvu-
lus), and other areas of ectopic gastric mucosa.
INFECTION AND INFLAMMATION
Q: Three of four Ga-67 photopeaks are acquired for
imaging.What are they and what is their abundance?
A: 185 kev (23% abundance), 300 keV (18%) and 394
keV (4%).The 91 kev (41%) is not acquired.
Q: Name the photopeaks of In-111 and their abun-
dance that are used for leukocyte imaging.
A: 173 kev (89%) and 247 keV (94%).
Q: Which collimator should be used for Ga-67 and
In-111?
A: Medium-energy collimator.A high-energy collimator
can be used, although its efficiency is less and image
acquisition time is longer.
Q: Image quality is not as good with Ga-67 as that
achieved with Tc-99m agents for which of the fol-
lowing reasons?
a. Decreased crystal sensitivity for high-energy
photons
b. Poor collimator efficiency
 Pearls, Pitfalls, and Frequently Asked Questions 559

Q: Which of the following statements is true regarding

c. Scatter and septal penetration from high-energy
In-111 oxine leukocytes?
photons
a. It is diagnostically useful for evaluating inflam-
d. Multiple photopeaks
matory lung disease.
e. Low administered dose
b. It has a high sensitivity for detecting osteo-
f. High background activity
myelitis of the spine.
A: All except d. However, if the multiple photopeaks are
c. It should not be used when the peripheral
not aligned correctly, image quality can be adversely
leukocyte count is less than 3,000/mm3.
affected.
d. It is the radiopharmaceutical of choice for intra-
abdominal infection.
Q: Pulmonary Ga-67 uptake is seen in which of the fol-
A: c and d are true.A minimal number of leukocytes
lowing diseases?
are needed for adequate radiolabeling and sensitiv-
a. tuberculosis
ity of the test. The lack of intra-abdominal clear-
b. histoplasmosis
ance makes it ideal for detecting intra-abdominal
c. sarcoidosis
infection.The false negative rate for osteomyelitis
d. Pneumocystis carinii infection
of the spine is high, in the range of 40%. It is insen-
e. Kaposi’s sarcoma
sitive and nonspecific for pulmonary inflamma-
f. cytomegalovirus
tory disease, although focal uptake noted should
g. lung cancer
be pursued diagnostically.
h. pneumonoconioses
A: All of the above except for Kaposi’s sarcoma, which
is not Ga-67 avid. Pearl
Tc-99m HMPAO is the preferred agent for localizing
Pearl infection in pediatric patients because of In-111 leuko-
cyte’s high radiation dose to the spleen, in the range of
Kaposi’s sarcoma is Tl-201 avid.
15–20 rads in the adult, but 30–50 rads in children.
Q: Ga-67 is taken up by the lungs due to drug toxicity.
Q: What is the optimal imaging time for In-111 labeled
Which drugs are the culprits?
leukocytes and Tc-99m HMPAO leukocytes?
A: Bleomycin is the most common. However uptake
A: In-111 labeled leukocytes are routinely imaged at
can be seen with cytoxin, nitrofurantoin, and amio-
24 hours. Imaging at 4–6 hours is less sensitive for
darone.
detection of infection.The one exception is inflam-
matory bowel disease in which imaging should be
Q: What is the role of Ga-67 in sarcoidosis?
done at 4 hours, because intraluminal shedding of
A: Ga-67 lung uptake is a sensitive test for the diagnosis
inflamed cells may result in inaccurate localization
of active alveolitis of sarcoidosis. Ga-67 may be
at 24 hours.Tc-99m HMPAO leukocytes should be
markedly increased in the setting of a normal chest
imaged at 1–2 hours for intra-abdominal infection
radiograph in early disease,and may be negative in the
because of biliary and renal clearance seen by
setting of an abnormal radiograph in inactive disease.
2 hours. Extra-abdominal infection can be imaged
later, usually 4 hours, allowing more time for back-
Pearl ground clearance.
Characteristic scintigraphic patterns of uptake of sar-
coidosis are: 1) the “panda sign,” due to uptake in the sali-
vary glands, parotids, and nasopharyngeal region; and 2) Pitfall
the “lambda sign,” due to paratracheal and hilar lymph Leukocytes may accumulate at site of inflammation with-
node uptake. out infection, such as intravenous catheters, nasogastric,
endogastric, and drainage tubes, tracheostomies, and
Q: Which leukocytes are labeled with In-111 oxine and colostomies. Leukocytes may accumulate at postopera-
Tc-99m HMPAO? tive surgical sites for 2–3 weeks and low-grade uptake
A: In-111 binds to neutrophils, lymphocytes, mono- may be seen at healing fracture sites.Accessory spleens
cytes, as well as erythrocytes and platelets.Tc-99m may be misinterpreted as infection and renal transplants
HMPAO preferentially binds to neutrophils. accumulate leukocytes.
560 NUCLEAR MEDICINE: THE REQUISITES
Pitfall
Intraluminal intestinal radioactivity can be the result of
swallowed or shedding cells from pharyngitis, sinusitis,
pneumonia or herpes esophagitis. Gastrointestinal bleed-
ing is another cause for false-positive intraluminal leuko-
cyte activity.
Pearl
Tc-99m fanolesomaba (NeutroSpec) is the newest
approved radiopharmaceutical for infection imaging. It
is a murine monoclonal antibody that binds to human
neutrophils. It was initially approved for acute appen-
dicitis; however, it will likely be used for osteomyelitis
and other infectious etiologies. It is not excreted intra-
abdominally. Its major advantage over In-111 oxine and
Tc-99m HMPAO leukocytes is that no blood handling is
required.
Pearl
F-18 FDG is expected to have a future role in infection
and inflammation imaging.
Q: Which of the following statements is not true?
a. A negative three-phase bone scan excludes
osteomyelitis with a high degree of certainty.
b. The specificity of the bone scan is poor in
patients with underlying bone disease such as
fractures, orthopedic hardware, and neuropathic
joints.
c. In-111 oxine and Tc-99m HMPAO leukocytes have
poor specificity for the diagnosis of osteomyelitis
in a patient with a hip prosthesis.
d. Because the three-phase bone scan may be
positive in a patient with a Charcot’s joint,
a radiolabeled leukocyte study should be per-
formed.
A: d is not correct.A radiolabeled leukocyte study may
also be a false positive. However, the combination of
a bone marrow study with a leukocyte study can be
diagnostic. Similarly hip and knee prostheses are
best evaluated with a bone marrow study as well. In
any clinical situation where the bone marrow distri-
bution may not be normal, a Tc-99m SC study is indi-
cated.
Pitfall
Although the three-phase bone scan has a high negative
predictive value for osteomyelitis in general, false nega-
tives have been reported in the neonate.
Q: Regarding osteomyelitis, which of these statements
is false?
a. The three-phase bone scan is a sensitive test for
diagnosis.
b. The three-phase positive scan is specific for
osteomyelitis.
c. A negative flow phase study almost always rules
it out.
d. In patients with prostheses, a bone marrow
study can be useful to rule out a false positive In-
111 leukocyte study.
A: b. The bone scan is the most sensitive test for
osteomyelitis;however,it is not specific.For example,
a three-phase bone scan can be seen with fracture, a
Charcot joint, tumor, and so forth.Although false-neg-
ative bone scans are rare, ischemia due to arterioscle-
rotic vascular disease can result in a false-negative
flow study.A bone marrow study in conjunction with
a labeled leukocyte study is the most accurate
method for diagnosing an infected prosthesis.
CENTRAL NERVOUS SYSTEM
Q: How is the diagnosis of brain death made?
A: The patient is in a deep coma with total absence of
brainstem reflexes and spontaneous respiration.
Reversible causes (e.g., drugs, hypothermia) must be
excluded; the cause of the dysfunction must be diag-
nosed (e.g., trauma, stroke); and the clinical findings
of brain death must be present for a defined period
of observation (6–24 hours). Confirmatory tests such
as electroencephalography (EEG) and radionuclide
imaging may be used to increase diagnostic certainty,
but the diagnosis is primarily clinical.The radionu-
clide study is more specific than EEG.
Q: Which radiopharmaceuticals are used to evaluate
brain death, and what are the advantages of each?
A: Tc-99m DTPA is inexpensive but more technically
demanding to use and interpret.Tc-99m HMPAO or
Tc-99m ECD are often preferred because no flow
study is required, just delayed images to visualize
radiotracer fixed in cortex.
Pearl
A “hot nose” may be seen on the flow-phase images and
delayed images as a result of shunting of blood from the
internal to the external carotid system that supplies the
face and nose in patients with severe carotid stenosis,
brain death, psychoactive drug use, and use of other
drugs that cause nasal congestion.

Q: What is the difference in mechanism of uptake
between fluorine-18 fluorodeoxyglucose (F-18
FDG) and the Tc-99m cerebral perfusion agents?
A: F-18 FDG is a glucose analog, and its uptake repre-
sents regional glucose metabolism. It is metaboli-
cally trapped intracellularly. Tc-99m HMPAO and
Tc-99m ECD are lipid-soluble cerebral perfusion
agents taken up in proportion to regional cerebral
blood flow. They fix intracellularly. In most cases,
cerebral blood flow follows metabolism.
Pearl
An example of a decoupling of metabolism and blood
flow is seen during the acute phase of a stroke. Blood
flow may be normal or increased for the initial 1–10 days
(luxury perfusion), but metabolism is decreased.
Q: How can single-photon emission computed tomog-
raphy (SPECT) brain perfusion or positron emission
tomography (PET) FDG imaging be useful in the dif-
ferential diagnosis of dementia?
A: Multi-infarct dementia is characterized by multiple
areas of past infarcts,recognized as areas of decreased
uptake that correspond to the vascular distribution
patterns as well as to changes in the deep structures
such as the basal ganglia and thalamus.Alzheimer’s
disease exhibits a characteristic pattern of bitemporal
and parietal hypoperfusion and hypometabolism.
Pick’s disease is associated with decreased frontal
lobe uptake. AIDS-dementia complex is associated
with a pattern of multifocal or patchy cortical regions
of decreased uptake, seen particularly in the frontal,
temporal,and parietal lobes and the basal ganglion.
Pearl
Although Alzheimer’s disease has a characteristic bitem-
poral-parietal pattern on perfusion imaging, it is often
not symmetrical. Decreased frontal lobe uptake may also
be seen.This pattern often cannot be differentiated from
the imaging pattern of Parkinson’s disease and Lewey
body disease, although they typically have very different
clinical presentations.
Q: What is the purpose of cerebral perfusion imaging
in patients with seizures? What is the expected PET
or SPECT pattern?
A: F-18 FDG PET or SPECT cerebral perfusion studies
can often localize the seizure focus in patients
requiring surgery (typically temporal lobectomy)
for seizure control. Interictally, a seizure focus
shows decreased metabolism on FDG PET and
decreased perfusion on SPECT; increased activity is
Pearls, Pitfalls, and Frequently Asked Questions 561
seen during a seizure (ictal). Normally, perfusion fol-
lows metabolism. In many surgical seizure centers,
depth electrodes are not required preoperatively if
the clinical picture, EEG, and SPECT study are all
consistent as to the location of the seizure focus.
Q: Which radiopharmaceuticals have been found use-
ful in imaging brain tumors,and what is their clinical
utility?
A: F-18 FDG PET imaging demonstrates increased
uptake in tumors owing to increased glycolysis.
Uptake of FDG is proportional to the malignant
grade of glioblastomas. PET determines tumor viabil-
ity after radiation therapy. SPECT with thallium-201
and Tc-99m sestamibi can be used in a similar man-
ner. Both T1-201 and PET FDG can differentiate lym-
phoma from infection, most often toxoplasmosis, in
AIDS patients. Uptake of T1-201 or FDG is indica-
tive of lymphoma.
Q: Name the radiopharmaceutical used for cisternogra-
phy and the most common clinical indication for
this study.
A: In-111 DTPA.The most common use of this radio-
pharmaceutical in modern practice is to confirm the
diagnosis of normal-pressure hydrocephalus, an
obstructive communicating form of hydrocephalus.
The next most common use is to localize cere-
brospinal fluid (CSF) leaks.
Pearl
The symptoms of normal pressure hydrocephalus are
incontinence, dementia, and gait disturbance.
Q: What is the characteristic pattern of normal pressure
hydrocephalus on radionuclide cisternography?
A: Persistent ventricular filling and evidence of a con-
vexity block.
CARDIAC
Pearl
Myocardial perfusion scintigraphy, whether performed
with SPECT or PET, is a “map” of relative blood flow to
viable myocardium.That is, for activity to be recorded in
the image,it must be delivered (blood flow) and taken up
by a myocardial cell (viable myocardium).
Q: How does the percent extraction of thallium-201
passing through the myocardial capillary bed com-
pare with the extraction of Tc-99m sestamibi and
Tc-99m teboroxime?
562 NUCLEAR MEDICINE: THE REQUISITES
A: Tl-201 has a myocardial extraction fraction of
approximately 0.85 in normal subjects at normal
flow rates.The myocardial extraction of Tc-99m ses-
tamibi and Tc-99m tetrofosmin is considerably
lower, 0.50 and 0.60, respectively.
Q: What percentage of Tl-201,Tc-99m sestamibi, and
Tc-99m tetrofosmin localizes in the heart?
A: For Tl-201, 3–4% of the administered dose localizes
in the heart in normal subjects; for Tc-99, sestamibi,
1.5%, and for tetrofosmin, 1.2%.
Q: What are the advantages and disadvantages of Tl-201
as perfusion agent for stress and rest myocardial
scintigraphy?
A: Advantages:Tl-201 requires only a single injection
because of redistribution; imaging can be per-
formed early after stress, within 10–15 minutes; it
can be used to assess viability. Disadvantages: its low
administered dose because of its high radiation
dosimetry; its poor imaging characteristics with
a low photopeak of 69 to 80 keV and high scatter
fraction; and its susceptibility to the effects of atten-
uation.
Pitfall
Patient motion can cause image degradation, create
a defect, or obscure a defect.
Q: What quality control should be routinely performed
to detect patient motion?
A: Review raw data in cinematic display. Review of the
sonogram can confirm the extent of the problem.
Pearl
The best method for correcting the problem of motion
is to repeat the study. If this is not possible, motion cor-
rection programs should be used. However, this software
only corrects for motion in the vertical axis. Motion in
the horizontal or diagonal axes will not be corrected.
Pitfall
Attenuation of photons by soft tissue can result in
decreased activity to the myocardium that might sug-
gest myocardial infarction if seen on both rest and
stress or as ischemia if only apparent on the stress
study. With females, breast attenuation results in
decreased activity of the anterior, septal, or lateral wall,
depending on their size and shape. Males characteristi-
cally have attenuation of the inferior wall, so-called
diaphragmatic attenuation.
Q: In what ways can the image interpreter determine if
fixed decreased activity is indeed pathological (i.e.,
infarction) or merely due to attenuation?
A: Review the raw data in the cinematic display to look
for soft attenuation. Review of the gated SPECT can
help determine if there is wall motion and thickening
that would indicate that it is not an infarct, but proba-
bly due to attenuation, not infarction.Attenuation cor-
rection programs can be helpful. Prone imaging has
been used to differentiate attenuation from infarction
in the inferior wall.
Pearl
To correct for attenuation, a transmission map must be
acquired. This has commonly been done by acquiring
transmission counts from a gamma source (e.g.,gadolinium-
153).With SPECT-CT systems, the acquired CT is used for
attenuation correction.
Q: What is the significance of lung uptake on Tl-201
exercise studies?
A: Lung uptake on exercise stress images, but not the
delayed images is consistent with exercise-induced
cardiac dysfunction.This finding is usually associ-
ated with three-vessel coronary artery disease.
Pearl
Lung uptake on Tl-201 images is easily perceived on pla-
nar images, but may not be appreciated in SPECT recon-
structed and reoriented images. Review of the cinematic
display of the raw projection data (low count planar
images at multiple angles around the patient) will nicely
show the lung uptake.
Q: What other scintigraphic finding suggest three-
vessel disease?
A: Exercise induced ischemic dilation. The normal
heart dilates during stress but gated SPECT is
acquired poststress when normal hearts have
returned to baseline size.
Pitfall
Incomplete normalization on delayed Tl-201 imaging
does not equate with a fixed perfusion defect. Insisting
on complete normalization before accepting an abnor-
mality as not “fixed” results in underdetection of ische-
mic areas versus scarred areas.
Q: What is the rationale for a second injection of Tl-201
versus simple delayed imaging to distinguish fixed
from reversible defects?

A: Relying solely on delayed imaging overestimates the
number of fixed myocardial defects. Redistribution
may take longer than the usual 3- to 4-hour delay and
may not even be complete by 24 hours.A low serum
Tl-201 level may not allow for significant redistribu-
tion. Reinjection of Tl-201 increases the serum level,
permitting further redistribution.
Pearl
Patients with left bundle branch block may have
reversible stress-induced hypoperfusion of the septum.
Patients with ischemia do not usually have isolated
septal involvement, but also apical and anterior wall
ischemia. This potential diagnostic problem can be
avoided by performing pharmacologic stress with
adenosine or dipyridamole.
Q: What is the relationship between the time after
myocardial infarction and the sensitivity of perfu-
sion imaging?
A: The sensitivity of perfusion imaging for detecting
defects caused by acute myocardial infarction is
greatest right after the infarct and diminishes with
time.This is different from “hot spot” imaging with
Tc-99m pyrophosphate, in which the greatest sensi-
tivity does not occur for a day or two after infarction.
Pitfall
Although myocardial perfusion scintigrams are positive
immediately after infarction, it is not possible to deter-
mine whether a given defect is new or old.A given cold
area may be caused by myocardial scar or acute myocar-
dial infarction.
Pearl
The primary cause of false negative exercise studies in
the diagnosis of coronary artery disease is failure to
achieve adequate exercise.
Pearl
After exercise, significant Tl-201 localization in the liver
usually indicates a poor exercise level.At peak exercise,
blood flow is diverted from the splanchnic circulation.
Pitfall
Quantitative analysis systems that rely on databases of
“normals” may not reflect the patient population in a dif-
ferent nuclear medicine department. Care must be taken
to not rely too heavily on these databases.
Pearls, Pitfalls, and Frequently Asked Questions 563
Q: What is the mechanism of action of dipyridamole?
A: Dipyridamole inhibits the action of adenosine
deaminase. By augmenting the effects of endoge-
nous adenosine, dipyridamole is a powerful vaso-
dilator.
Q: What effect can a cup of coffee have on a dipyri-
damole or adenosine stress test?
A: Caffeine in coffee, tea, soft drinks, or foods such as
chocolate are chemically related to dipyridamole
and adenosine and can block the effect of dipyri-
damole pharmacological stress testing.
Q: What percentage of stenosis at rest is necessary in
the coronary arteries for resting blood flow to be
affected?
A: Coronary artery stenosis greater than 85–90% is
required before flow is diminished at rest. Not all
stenoses are created equal. Long irregular stenotic
segments have more effect than discrete short-
segment stenoses.
Q: Why is imaging delayed for 30–90 minutes after
administration of Tc-99m sestamibi or Tc-99m tetro-
fosmin?
A: Although myocardial uptake is rapid with both
Tc-99m sestamibi and Tc-99m tetrofosmin, lung and
liver uptake are also significant.The lung and liver
clear with time and the target-to-background ratio
improves.
Q: What pharmaceutical is administered that allows the
Tc-99m to bind to the red blood cell?
A: Stannous (tin) pyrophosphate is the usual agent.
Stannous chloride has been used.
Q: To what part of the red blood cell does the Tc-99m
label bind?
A: Tc-99m binds to the beta chain of hemoglobin when
the patient is pretreated with stannous ion.
Pitfall
Injection of labeling materials through a heparinized
intravenous line can significantly decrease the yield with
in vivo and modified in vivo RBC labeling.
Pearl
The in vitro kit method (UltraTag) of radiolabeling red
blood cells is the present day preferable methodology
because it has the highest labeling efficiency, greater
than 97%. This results in less background and higher
accuracy.
564 NUCLEAR MEDICINE: THE REQUISITES
Q: What are the considerations for selecting the num-
ber of frames in a gated blood pool study?
A: Selecting the number of frames to divide the cardiac
cycle is a balance between having enough frames to
capture the peaks and valleys of the ventricular
time–activity curve versus the need to acquire a sta-
tistically valid number of counts in each frame. For
gated ventriculography (MUGA), using 16 frames
achieves this compromise. For gated SPECT myocar-
dial perfusion imaging, 8 frames is the usual compro-
mise.Too few frames will “average out”the peaks and
valleys.Too many frames increases the imaging time
required for a given number of counts per frame.
Pitfall
In calculation of the left ventricular ejection fraction, too
high an estimate of the background counts per pixel will
result in a falsely high ejection fraction.This can happen
if the background area includes activity from the spleen.
Pearl
Variations in the length of the cardiac cycle can be recog-
nized on gated perfusion or blood pool studies if the
time–activity curve trails off or fails to approximate the
height of the initial part of the curve. Significant asym-
metry (>10%) of the height of the curve at the beginning
and the end may indicate significant arrhythmia.
Q: What do amplitude and phase images portray?
A: Amplitude and phase images are parametric or
derived images.The amplitude image portrays the
maximum count difference at each pixel location
during the cardiac cycle.High ejection fraction areas
have high amplitude, and background areas have
low amplitude.The phase image portrays the timing
of cyclical activity with respect to a reference stan-
dard, usually the R wave.
Q: What is the hallmark of a ventricular apical aneurysm
by phase analysis?
A: Aneurysms demonstrate paradoxical motion. Activity
in the area of the aneurysm is typically 180 degrees
out of phase with the rest of the ventricle.
PULMONARY
Q: What are the two most commonly used radiophar-
maceuticals for ventilation imaging? What are their
advantages and disadvantages?
A: Xenon-133 and Tc-99m DTPA aerosol. Xenon-133
demonstrates more clearly the physiology of respira-
tion and is very sensitive to detection of obstructive
airway disease manifested by slow washout.The dis-
advantage is the rapid washout, limiting the views
obtainable, its suboptimal image quality due to the
low photopeak (81 keV) and poor count rate image.
Tc-99m DTPA aerosol allows high count images in
all projections; however, the images are comparable
only to the inspiratory phase of xenon-133. With
obstructive airway disease, particles impact in the
proximal bronchi, potentially causing interpretation
difficulties.
Pitfall
Xenon-133 will be taken up and cleared slowly from liv-
ers with fatty metamorphosis.This should not be con-
fused with pulmonary delayed washout.
Q: What is the minimum number of particles recom-
mended for pulmonary perfusion imaging?
A: Pulmonary perfusion scanning assumes a statisti-
cally even distribution of particles throughout the
lung.This requires at least 100,000 particles in nor-
mal adults, and 200,000–500,000 particles are gener-
ally administered.
Q: How should the dose of technetium-99m macroag-
gregated albumin (MAA) be adjusted in pediatric
patients?
A: Radiopharmaceutical doses are always adjusted with
respect to patient size or age in the pediatric popu-
lation. With Tc-99m MAA it is also necessary to
reduce the number of particles.
Q: What is the size range of MAA particles?
A: In commercial preparations the majority of particles
are 20–40 μm, with a range of 10–90 μm.
Pitfall
Withdrawing blood into a syringe with Tc-99m MAA par-
ticles may create a small radioactive embolus that shows
up as a “hot spot”on subsequent images.
Pitfall
Failure to resuspend the Tc-99m MAA particles before
administration may result in clumping of particles together
and the presence of “hot spots”on subsequent imaging.
Q: What is the biological fate of MAA particles?
A: MAA particles are physically broken down in the
lung. Delayed imaging performed several hours
after pharmaceutical administration demonstrates

activity in the reticuloendothelial system because of
phagocytosis of the breakdown particles.
Pearl
One way to determine whether radioactivity outside of
the lungs is caused by free Tc-99m or right to left shunt-
ing of Tc-99m MAA is to image the brain. Free pertechne-
tate should not localize in the brain, but rather in the
thyroid, salivary glands, and stomach, whereas Tc-99m
MAA particles that gain access to the systemic circulation
will lodge in the first capillary bed that they encounter,
including the capillary bed in the brain. If no brain
uptake is seen, there is no significant right to left shunt.
Q: What is the preferred patient position during admin-
istration of Tc-99m MAA?
A: Administering Tc-99m MAA with the patient supine
results in a more homogeneous distribution of parti-
cles in the lung than when the patient is sitting or
standing. Gravitational effects result in more basilar
distribution when injection is accomplished with
the patient upright.
Pitfall
In lateral views of the lung obtained for a fixed number
of counts,“shine-through”from the contralateral lung can
give the false impression of activity arising from the side
being imaged.This is most dramatically demonstrated in
patients who after pneumonectomy show no activity on
anterior or posterior views but a near-normal appearance
can be seen because of the shine-through phenomenon.
Pitfall
In analysis of perfusion scintigrams, failure to recognize
the significance of decreased versus absent activity is
a potential pitfall. Not every clot is 100% occlusive of the
circulation. Significantly diminished activity needs to be
recognized as one of the patterns caused by pulmonary
emboli.
Pitfall
In some patients with fatty liver, retained activity in the
liver on Xe-133 scans can be confused with retained
activity or delayed washout at the right base. Remember
that xenon is fat soluble and will show significant accu-
mulation in patients with fatty liver.
Pitfall
The pulmonary hili are photon-deficient structures
caused by the displacement of lung parenchyma by large
Pearls, Pitfalls, and Frequently Asked Questions 565
vascular and bronchial structures. Failure to remember
this can result in false positive interpretations, especially
for defects seen on posterior oblique images.
Pitfall
If the patient is placed supine for V/Q imaging but the
chest radiograph was obtained with the patient
upright, it can be difficult to correlate findings on the
examinations. For example, free fluid may collect in
a subpulmonic location or obscure the lung base in the
upright position.With the patient supine, the fluid may
layer out posteriorly or collect in the fissures.Also, the
apparent height of the lungs may be different, as may
the heart size. Ideally, imaging studies should be per-
formed with the patient in the same position for all
examinations. On the other hand, if there is significant
pleural fluid, it may be desirable to image the patient in
more than one position to prove that a defect is caused
by mobile fluid.
Q: What is the stripe sign?
A: The stripe sign refers to a stripe or zone of activity
seen between a perfusion defect and the closest
pleural surface. Because pulmonary emboli are typi-
cally pleura based, the stripe sign suggests another
diagnosis, often emphysema. Rarely, in the resolution
of pulmonary emboli, a stripe sign develops as circu-
lation is restored.
Q: What is the physiological basis for perfusion defects
in areas of poor ventilation?
A: The classic response to hypoxia at the alveolar level
is vasoconstriction. Shunting of blood away from the
hypoxic lung zone maintains oxygen saturation.
Q: What is the shrunken lung sign?
A: The lungs may appear smaller than usual in patients
sustaining multiple small emboli, such as fat emboli,
that distribute uniformly around the lung periph-
ery.
Q: What is the classic appearance of multiple pul-
monary emboli on lung perfusion scintigraphy?
A: Multiple pleura-based, wedge-shaped areas of signifi-
cantly diminished or absent perfusion.The size of
the defects may vary from subsegmental to segmen-
tal or may involve an entire lobe or lung.
Q: What are the most common clinical signs and
symptoms in patients with confirmed pulmonary
embolism?
A: In the PIOPED study, the three most common pre-
senting symptoms (and approximate percentage fre-
quency) were dyspnea, 80%; pleuritic chest pain,
566 NUCLEAR MEDICINE: THE REQUISITES
60%;and cough,40%.Hemoptysis (15%) and leg pain
(25–30%) were less common. On physical examina-
tion, lung crackles (60%) were encountered much
more often than leg swelling (30%) or pleural fric-
tion rub (5%).Both the heart rate and the respiratory
rate were elevated on average in the PIOPED study
in patients with pulmonary embolism.
Q: What is the sensitivity of the high-probability scan
category for detecting pulmonary embolism?
A: In the PIOPED study, 41% of patients with pul-
monary embolism had a high-probability scinti-
graphic pattern.Thus the majority of patients with
pulmonary emboli have intermediate or low-proba-
bility scans.
 Index

Note: Page numbers followed by f refer to figures; page numbers followed by t refer to tables;
page numbers followed by b refer to boxes.

A Acute pyelonephritis, 255

Acute testicular torsion
Angstroms, 21b
Anne Arbor staging system
Abdominal imaging scrotal scintigraphy for, 260 for HD, 270b
for infections, 418 views of, 260f for lymphomas, 336b
for tumors, 272–273 Adenosine Antineutrino, 25
Acalculous biliary disease. See Chronic chemical structure of, 465f Antrum
acalculous cholecystitis pharmacologic effect of, 461–462 anatomy of, 356f
Accelerated acute graft rejection, 244 side effects of, 462–465, 465t gastric motility and, 355–356
ACE. See Angiotensin converting enzyme stress test protocol with, 465b Appendicitis, 412
ACE inhibition renography, 230–234 Adrenocortical scintigraphy, 105–109 Arterial stenosis, 246, 248
criteria for, 237b androgen excess and, 108–109 Arterial thrombosis, 246
image interpretation of, 232–233 Cushing’s syndrome and, 106–108, 108f Atomic mass unit
imaging protocol for, 233, 235 hyperaldosteronism and, 108 formula for, 21b
indications for, 230 incidentalomas and, 109 mathematic definition of, 23
protocol for, 233b normal, 106 Atoms
renogram pattern with, 237f protocol for, 106b Bohr model of, 21–22, 21f
reporting of, 233–234 radiopharmaceuticals for, 105–106 mass-energy equivalence of, 23
Achalasia suppression studies with, 106 structure of, 20–21
definition of, 557 Adrenomedullary scintigraphy, 109–112 Auger electron
manifestation of, 347 clinical applications of definition of, 22b
semisolid meal and, 351f–352f neuroblastoma and, 111–112 in radionuclide decay, 28
Acquired immunodeficiency syndrome pheochromocytoma and, 111 Avogadro’s number, 21b
(AIDS), 431 drug interference with, 110, 110t
Activity-induced enthesopathy, 139 normal, 111
Acute acalculous cholecystitis, 174–175 radiopharmaceuticals for, 110 B
cholescintigraphy for, 174–175, 174t technique of, 110
conditions associated with, 174b Adult respiratory distress syndrome, 535–536 Background radiation, 40
Acute cholecystitis, 168–174 AIDS. See Acquired immunodeficiency Backscatter peak, 38
blood flow and, 548 syndrome Barium esophagography, 351
cholescintigraphy for, 170–174 AIDS-dementia complex, 431 Barium follow-through study, 382
accuracy of, 170b, 171–172, 173t Akinesis, 486 Barrett’s esophagus, 380
false positives in, 173b Alcoholic liver disease, 197, 200 Basophils, 389
rim sign in, 173–174, 174f, 548 Alpha decay, 24–25 Becquerels, 28
clinical presentation of, 168 Alpha particles, 542 Bell-clapper deformity, 259, 553
pathophysiology of, 168, 169b Alzheimer’s disease Bernstein acid infusion test, 351
ultrasonography for, 169–170, 170b imaging of, 428, 430–431 Beta minus decay, 25
Acute epididymitis bitemporal-parietal patterns and, 561 Beta particles, 30, 542
scrotal scintigraphy for, 260 PET, 429f, 430f Biliary atresia, 180–182
views of, 262f SPECT, 429f–430f cholescintigraphy for, 550
Acute graft rejection incidence of, 428 detection of, 549
dynamic renal scintigraphy for, 244, 251–252 Angiotensin converting enzyme (ACE) Biliary diversion surgery, 186
time course of, 249f–250f physiology of, 232f Biliary duct obstruction, 177–180
time-activity curves of, 248f, 250f renography with, 230–234 biliary-to-bowl transit and, 178, 179b, 180f
views of, 247f–248f renovascular hypertension and, 230 clinical presentation of, 177t

567
568 INDEX
Biliary duct obstruction (Continued)
delayed transit and, 549
high-grade, 177, 178f
imaging of
cholescintigraphy for, 177–180
MRCP for, 177
ultrasonography for, 177
partial, 177, 179f
pathophysiology of, 177
postoperative, 182, 184
Biliary dysfunction, postoperative, 182–190
biliary diversion surgery and, 186
biliary leaks and, 185–186
liver transplants and, 187
other gastrointestinal surgical procedures
and, 186–187
partial biliary obstruction, 182, 184
postcholecystectomy pain syndrome and,
182
SOD and, 184–185
protocol for, 186b
sequential images of, 187f
tumors and
differential diagnosis of, 189t
focal nodular hyperplasia and, 187, 189
hepatic adenomas and, 189
hepatocellular carcinoma and, 189–190
Biliary dyskinesia. See Sphincter of Oddi
dysfunction
Biliary leaks, 185–186
Biliary spasm. See Sphincter of Oddi
dysfunction
Binding energy, 542
Biological half-life
mathematics of, 30
SPECT and, 544
of Tc-99m HMPAO, 395
Bladder carcinoma, 344–345
Bladder volume, 553
Blood urea nitrogen (BUN) test, 252
Bohr model of atoms, 21–22, 21f
Bone dysplasias, 129
Bone marrow scintigraphy, 152–153
Bone mineral measurement, 152–158
applications of, 155
classification of, 157
DPA for, 154
DXA for, 154
SPA for, 153–154
Bone pain, 297–300
Bone scan
frequently asked questions regarding,
546–548
with Ga-67, 406
nonspecificity of, 547
for osteomyelitis, 405–407
with radiolabeled leukocytes, 406–407
Bone tumors, 345
Brain death, 560
Brain death scintigraphy, 435–438
images from, 427f–430f
interpretation of, 436, 438
methodology of, 435, 439b
radiopharmaceuticals for, 435–436
Brain imaging. See Cerebral perfusion
imaging; Cisternography
Breast cancer
diagnosis of, 338, 340
lymph node evaluation and, 340–341
lymphoscintigraphy for, 301
mammography for, 277–278
osseous metastases and, 120–121
PET for, 341–342
Breast cancer (Continued ) Cavernous hemangiomas (Continued)
primary cancer detection and, 340 dosimetry of, 192, 192t
skeletal scintigraphy and, 124–125 image interpretation with, 192–193
tumor imaging for, 277–278 methodology of, 192, 192b
ultrasonography for, 277 pharmacokinetics and, 191
Brodie’s abscess, 404 CEA-Scan
Bronchoalveolar lavage, 416 accuracy of, 288–289
Budd-Chiari syndrome, 197 characteristics of, 288t
BUN. See Blood urea nitrogen test dosimetry of, 289, 291t
Butterworth filter, 58–59, 58f indications for, 287–288
pharmacokinetics of, 287
Cellulitis, 154b
C Centimeter-gram-second (CGS), 28t
Cerebral anatomy, 419–422
C3a, 385t arterial, 421f
C5a, 385t brain functions and, 420f
Caffeine, 465f CSF flow dynamics and, 442f
Calories, 21b lobar, 420f
Cancer, 263–301. See also Peptide receptor perfusion and, 422f
imaging;Tumor imaging; specific types venous, 422f
frequently asked questions regarding, Cerebral perfusion imaging, 419–442
553–557 brain anatomy and, 432–435
lymphoscintigraphy for, 300–301 arterial, 421f
palliation for, 297–300 brain functions and, 420f
peptide radiotherapy for, 284 lobar, 420f
PET for, 304 perfusion and, 422f
benign variants in, 307–313 venous, 422f
malignant patterns in, 313–314 clinical applications of, 427–442
normal distribution and, 306–307 cerebrovascular disease and, 431–438
patient scheduling concerns with, 308b dementia and, 427–430
protocol for, 305–306, 305b epilepsy and, 438–439
radiopharmaceuticals for, 303–305 head trauma and, 442
Captopril, dosage of, 552 Huntington’s chorea and, 442
Carbon-11, 5t movement disorders and, 441–442
Cardiac PET, 476–481 psychiatric disorders and, 442
coronary artery disease and, 478–479 stroke and, 431–438
myocardial viability and, 479–480 tumor imaging and, 439, 441
protocol for, 480–481, 480b indications for, 425b
radiopharmaceuticals for, 476–479, 477t methodology of, 426–427
FDG and, 479 PET, 427b
N-13 and, 477 SPECT, 427b
O-15 water and, 478 radiopharmaceuticals for
Rb-82 and, 477–478 background of, 420, 423
Tc-99m MIBI and, 480–481 dosimetry of, 428t
Cardiac radiopharmaceuticals, investigational, FDG and, 426
506 normal distribution of, 426
Cardiac stress tests, 458–459 PET and, 425b, 425t
discontinuation of, 464b Tc-99m and, 425–426
drug interference with, 463b Cerebral spinal fluid (CSF)
exercise, 459–461 flow dynamics of, 442f
exercise and, 459–461 flow patterns of, 444t
indications for, 463b leaks of, 445–449
methods of, 464, 464b pledget placement for, 447f
perfusion scintigraphy and, 459 protocol for, 447b
pharmacologic, 461–466 Cerebrovascular disease
adenosine and, 461–465 brain death scintigraphy and, 435–438
comparison of, 465t images from, 437f–438f
dipyridamole for, 461–465 interpretation of, 436, 438
dobutamine for, 465–466 methodology of, 436, 439b
scintigraphic, 459 radiopharmaceuticals for, 435–436
Cardiac toxicity, 502 PET for, 431–435
Cardiomyopathy, 502 SPECT for, 431–435
Cardiovascular disease, 413–414 CGS. See Centimeter-gram-second
Cavernous hemangiomas, 190–195 Characteristic x-rays, 28
diagnostic imaging of, 550 Chemokines, 385t
computed tomography for, 191 Chemotactic peptides, 401
false-positives in, 550 Chemotaxis, 385t
magnetic resonance imaging for, 191 Chemotherapy, 550–551
pathology of, 190–191 Chest pain
ultrasonography for, 191 in emergency rooms, 482
Tc-99m-labeled RBC scintigraphy for, risk stratification of, 484
191–195 triage of, 483f
accuracy of, 193–195 Child abuse, 135, 137

Cholangiocarcinoma, 332–333
Choledochal cysts
cholescintigraphy of, 180
schematics of, 182f
Cholescintigraphy, 160–190
cholecystokinin and, 163–166
gastrointestinal actions of, 164b
methodology for, 164–166
physiology of, 163–164, 164f
sincalide and, 164, 164b, 165b
clinical applications for, 168–190
acute acalculous cholecystitis and,
174–175
acute cholecystitis and, 168–174
biliary atresia and, 180–182
biliary duct obstruction and, 177–180
choledochal cysts and, 180
chronic cholecystitis and, 175–177
postoperative biliary tracts and, 182–190
clinical indications for, 160b
false-positives in, 548
methodology of, 163, 163b
food and, 548
morphine augmentation and, 172–173,
549
normal
biliary clearance and, 168
blood flow and, 166
gallbladder filling and, 167
hepatic function and, 166–167
liver morphology and, 166
patient history and, 162–163
patient preparation for, 163
radiopharmaceuticals for, 160–162, 160t
dosimetry of, 162, 163t
kit preparation of, 161
pharmacokinetics of, 162
Tc-99m chemistry and, 160–161, 161f
uptake mechanisms of, 161–162
Chromatography, 9
Chronic acalculous cholecystitis, 175–176
Chronic allograft nephropathy, 244–245
Chronic cholecystitis, 175–176
cholelithiasis and, 175
cholescintigraphy for, 175–176
delayed visualization of, 549
Chronic renal failure, 230f
Cimetidine, 376
Cirrhosis, 207f
Cisternography, 442–449
clinical applications of, 443–449
CSF leaks and, 446–449
hydrocephalus and, 443–444
surgical shunt patency and, 444–445
interpretation of, 444, 444t
methodology of, 442, 443b
radiopharmaceuticals for, 442
dosimetry of, 443, 443t
pharmacokinetics of, 442–443
Clotting system, 385t
Cobalt-57, 5t
Coincidence peaks, 38, 545
Collimators
converging-hole, 42–43, 43f
diverging-hole, 42–43, 43f
high resolution, 42f
parallel-hole, 41–42
pinhole, 41, 41f
purpose of, 544
specialty, 43
for SPECT, 53–54
types of, 544
Colloid nodular goiter, 93
Colorectal cancer, 286–290
immunoscintigraphy for
accuracy of, 288–289
adverse effects from, 290
methodology of, 290b
radiopharmaceuticals for, 287–288
PET for, 331
scintigraphic view of, 206f
staging of
Dukes’ system for, 287b, 333b
TNM classification for, 333b
Complex regional pain syndrome, 137–138
Compton edge, 38
Compton effect, 32
Compton plateau, 38
Compton scatter
bad photons from, 40
gamma ray spectrometry and, 38
mass-energy interactions and, 32, 32f
photon energy and, 542
scintigraphic image quality and, 543
Compton valley, 38
Computed tomography pulmonary
arteriography (CTPA), 534–535
Computers
data analysis with, 50
data display with, 50–51
digital image creation with, 49–50
Congenital heart disease, 503
Contrast angiography, 364, 558
Conversion electron, 22b
Coronary artery disease, 457–485
cardiac stress tests and, 458–459
discontinuation of, 464b
drug interference with, 463b
exercise, 459–461
indications for, 463b
methods of, 464
pharmacologic, 461–466
scintigraphic, 459
image interpretation for, 466–480
attenuation and, 469–470
coronary anatomy and, 466–467, 467f
extracardiac uptake and, 470
infarction and, 470, 474
ischemia and, 470, 474
left bundle branch block and, 474
multiple perfusion defects and, 474–477
myocardial perfusion and, 466–467
myocardial viability assessment and,
477–480
quality control and, 467–468
rest v. stress and, 468–469
imaging protocols for, 466, 481–482
PET for, 487–488
physiology of ischemia and, 458
prognosis of, 482–485
acute ischemic syndromes and, 482–487
chest pain and, 482, 483f, 484f
chronic ischemic syndrome and, 483,
485
sarcoidosis and, 485
quantitative analysis of, 480–481
RVG for, 501
Cortical cerebral scintigraphy. See Cerebral
perfusion imaging
Cr-51, 2
Creatinine, 252
Crohn’s disease, 413
CRP, 385t
CSF. See Cerebral spinal fluid
CTPA. See Computed tomography pulmonary
arteriography
INDEX 569
Curie, 28
Cushing’s syndrome
adrenocortical scintigraphy and, 106–108
diagnostic pattern of, 108f
Cutoff filter, 58
Cystic duct syndrome. See Chronic
acalculous cholecystitis
Cytokines
infection scintigraphy with, 401
inflammation and, 385t
Cytomegalovirus, 417
D
Decay constant, 28–30, 542
Deep vein thrombosis, 536–539
Dehydration, 551
Delayed graft function. See Vasomotor
nephropathy
Delayed testicular torsion
scrotal scintigraphy for, 260
views of, 261f
Dementia, 427–428
causes of, 428b
differential diagnosis of, 561
frontotemporal, 431, 432f
HIV/AIDS-related, 431
posterior, 428–431
vascular, 431, 433f
Diabetic foot
infection scintigraphy for, 407–408
osteomyelitis of, 407f
Diabetic gastroparesis, 358
Diffuse esophageal spasm, 348
Digital images
histogram creation of, 49–50
list mode creation of, 49, 50f
Diphosphonate, 114f
Dipyridamole
chemical structure of, 465f
pharmacologic effect of, 461–462
side effects of, 462–465, 465t
stress test protocol with, 464b
Diuretic renography, 234, 237–239
accuracy of, 552
conditions studied by, 238b
interpretation of, 239
limitations of, 239
methodology of, 237–239
diuretic administration for, 238–239
patient preparation for, 237–238
protocol for, 238b
radiopharmaceuticals for, 237
time-activity curves of, 241f
Dobutamine
chemical structure of, 465f
pharmacologic mechanisms of, 465
Dopamine neurons, 442f
Dose calibrators, 17–18
DPA. See Dual-photon absorptiometry
Dual energy x-ray absorptiometry (DXA), 154
Dual-photon absorptiometry (DPA), 154
Dukes’ staging system, 287b, 333b
DXA. See Dual energy x-ray absorptiometry
Dynamic renal scintigraphy, 223–252
clinical applications of
renal transplant evaluation and, 239–252
renovascular hypertension and,
230–234
urinary tract obstruction and, 234–239
computer processing of, 226–229
differential function and, 228–229
570 INDEX
Dynamic renal scintigraphy (Continued )
dynamic renal function time-activity
curve and, 227–228
normal renogram and, 228f
perfusion time-activity curve and, 227
semiquantitative renal function assess-
ment and, 228
interpretation of
clearance phase and, 229–230
cortical function phase and, 229
flow phase and, 229, 229f
methodology of
dose and, 226
image acquisition and, 226
patient positioning and, 226
preparation and, 226
protocol for, 227b
Dyskinesis, 495
E infection scintigraphy of, 418
Ectopic thyroid tissue, 93–94
Effective half-life, 30
Effective renal plasma flow (ERPF)
cameras and, 254–255, 255f
frequently asked questions regarding, 551
kidney function and, 216
radiopharmaceuticals used for, 219b, 253
technique for, 252–253
Egg sandwich, 359
Ejection fraction, 497–499
calculation of, 498f
formula for, 497
RVG and, 497
Electromagnetic radiation, 22–23
energy spectrum of, 22f
mathematics of, 22–23
Electron capture, 26
Electron volt, 21b
Electrons
characteristics of, 22b
energy equivalence of, 23t
rest mass of, 21b
Elementary charge, 21b
Eluate contaminants, 7–8, 541
Enchondromas, 129
Energy, 23
Enteropathy, 383
Eosinophils, 389
Epilepsy, 438–439, 561
Equilibrium gated blood pool
ventriculography, 491–495
acquisition techniques for, 492–495, 493f
functional parameters determined by, 499b
protocol for, 495b
radiopharmaceuticals for, 491
ERPF. See Effective renal plasma flow
Esophageal carcinoma, 326–330
diagnosis of, 328
restaging of, 330
staging of, 328, 330
therapy response monitoring for, 330
Esophageal endoscopy, 351
Esophageal motility, 347–350
classification of, 347b
motor disorders of
achalasia and, 347
diffuse esophageal spasm and, 348
nutcracker esophagus and, 348
other, 348
scleroderma and, 347–348
scintigraphy for, 348–350
Esophageal motility (Continued) Gallium-67 (Continued)
accuracy of, 350 production of, 4
analysis of, 349 tumor imaging with, 264–273, 555
dosimetry of, 348 clinical applications of, 269–273
methodology of, 349 interpretation of, 267–268
radiopharmaceuticals for, 348 methodology of, 266–267
Esophagus tumor uptake mechanisms of, 554–555
anatomy of, 347f Gallium-68, 5t
functions of, 347, 347f Gamma ray spectrometry
Euler’s number, 21b backscatter peak in, 38
Exercise stress tests, 459–461 coincidence peaks in, 38
endpoints for, 463b Compton scatter in, 38
treadmill for, 464b Compton valley, edge, and plateau in, 38
iodide escape peak in, 38
lead characteristic x-ray peak in, 38
F photopeak in, 37–38
Gamma rays
Fatty acid radiopharmaceuticals, 506 Compton effect with, 32
FDG. See Fluorine-18 fluorodeoxyglucose Compton scattering with, 32, 32f
Fever definition of, 23, 542
pair production with, 31
uveoparotid, 416 photoelectric absorption with, 31–32, 31f
Fibrin, 385t in radionuclide decay, 27–28
Fibroblasts, 385t Gamma scintillation camera, 39–49
Filters, 58–59, 544 bad photons in, 544
Fine needle aspiration, 89 background-produced, 40
Fission moly, 6 Compton scatter-produced, 40, 543
Fissure sign, 527, 533f patient-produced, 39–40, 40f
Flare phenomenon, 547 characteristics of, 44–45
Fluorine-18 fluorodeoxyglucose (FDG) clinical use of, 48
chemistry of, 304 collimators in
decay scheme of, 26f converging-hole, 42–43, 43f
distribution of, 304 diverging-hole, 42–43, 43f
infection scintigraphy with, 399 high resolution, 42f
physical characteristics of, 5t parallel-hole, 41–42
production of, 4 pinhole, 41, 41f
thyroid studies with, 95–96 specialty, 43
uptake of energy window of, 543
factors affecting, 308b event localization with, 43–44
mechanisms of, 304 gamma ray detection in, 43
Focal nodular hyperplasia, 187, 189, 550 image recording with, 44
Foods, 76 quality control for
Fourier phase analysis, 499–500 field uniformity in, 46–47
Fracture detection homogenous flood field and, 543
FDG uptake in, 313f linearity in, 46–47
skeletal scintigraphy of, 133–135 spatial resolution in, 47–48
Fundus summary of, 47b
anatomy of, 356f schematic of, 34f
gastric motility and, 355 signal processing with, 43–44
window setting for, 48–49
Gastric emptying scintigraphy, 358–362
G emptying rate and, 557
frequently asked questions regarding,
Ga-67. See Gallium-67 557–558
Gallbladder contraction, 549 indications for, 359, 359b
Gallbladder spasm. See Chronic acalculous interpretation of
cholecystitis decay correction and, 361
Gallium-67 (Ga-67) geometric mean attenuation correction
chemistry of, 264, 386 for, 361
distribution of left anterior oblique attenuation correc-
factors affecting, 266b tion for, 361
normal, 265f quantitative analysis for, 362
dosimetry of, 266, 267t scatter correction for, 362
infection scintigraphy with, 386–388 methodology of, 359–362
distribution in, 387 image acquisition frequency and,
dosimetry in, 388, 393t 361–362
imaging characteristics of, 387 liquid emptying and, 362
methodology of, 387–388, 392b meal and, 359
pulmonary infections and, 414–416 protocol for, 360t
uptake mechanisms in, 386 single v. dual isotope and, 359, 361
infection seeking time of, 396b solid emptying and, 362
pharmacokinetics and, 264–266, 387 solid v. liquid meal and, 359
physical characteristics of, 5t, 13, 265t, 387t study length and, 361

Gastric emptying scintigraphy (Continued )
radiopharmaceuticals for, 358
therapeutic effectiveness evaluation with,
362
Gastric motility, 354–362
gastric stasis syndromes of, 357–358
causes of, 358b
diabetic gastroparesis and, 358
nonulcer dyspepsia and, 358
pharmacological therapy and, 357–358,
358b
physiology and, 355–357
anatomy and, 356f
antrum and, 355–356
fundus and, 355
gastric emptying and, 357b, 357f
lag phase and, 356–357
motilin and, 357
scintigraphy for, 358–362
indications for, 359, 359b
interpretation of, 361–362
methodology of, 359–362
radiopharmaceuticals for, 358
therapeutic effectiveness evaluation
with, 362
Gastroesophageal reflux, 350–354
diagnosis of, 351
salivagram for, 354
scintigraphy for, 351–354
accuracy of, 354
interpretation of, 352–354
methodology of, 352
Gastroesophageal scintigraphy, 348–350
accuracy of, 350
analysis of, 349
dosimetry of, 348
adults and, 348t
children and, 348t
frequently asked questions regarding,
557–558
methodology of, 349
esophageal transit protocols for, 349b
semisolid meal protocol for, 349b
radiopharmaceuticals for, 348
Gastrointestinal bleeding, 364–372
contrast angiography for, 364
gastrointestinal tract vascular supply and,
374f
Tc-99m RBC scintigraphy for, 366–372
accuracy of, 371–372
cell labeling for, 366–368, 366b, 367b
diagnostic criteria of, 369–370
dosimetry of, 380, 380t
effectiveness of, 372
interpretation of, 368–369
methodology of, 368, 368b
physical characteristics of, 379t
pitfalls of, 370–371, 376b
Tc-99m SC scintigraphy for, 364–366
dosimetry of, 368, 368t
effectiveness of, 372
interpretation of, 365–366
methodology for, 365, 365b
physical characteristics of, 379t
time-activity curves of, 365f
views of, 365f, 366f
Gastrointestinal duplications, 379
Gastrointestinal stromal tumor (GIST)
images of, 336f
PET for, 333
Gastrointestinal tumors, 332–333
Gastrointestinal vascular supply, 374f
Geiger-Müller counters, 35
GFR. See Glomerular filtration rate
GIST. See Gastrointestinal stromal tumor
Glomerular filtration rate (GFR)
agents used for, 219b
cameras for, 254–255, 255f
frequently asked questions regarding, 551
normal values of, 254
radiopharmaceuticals for, 254
technique for, 252, 254
Glucagon, 377
Goiter. See also Multinodular toxic goiter
conditions associated with, 92b
radioiodine therapy for, 101
thyroid scintigraphy and, 92–93
types of, 93
Graves’ disease
pharmacologic therapy for, 99
radioiodine therapy for, 100–101
surgical therapy for, 99
thyroid uptake studies and, 85
H Hyperparathyroidism
Half-life
biological, 30
decay constant and, 542
effective, 30
mathematics of, 28–30
parent-daughter relationships and, 540
Hamming filter, 58, 58f
Hann filter, 58
Hanning filter, 58
Hashimoto’s disease, 86
Hashitoxicosis, 86
HD. See Hodgkin’s disease
Head and neck cancers
nodal classification systems for, 329b
PET for, 324–325
indications of, 328b
staging with, 329f, 328f
prognosis of, 324
tumor imaging for, 272
Helicobacter pylori
ulcers and, 362
urea breath test for, 363
Hemangiomas. See Cavernous hemangiomas;
Liver hemangiomas
Hepatic adenomas, 189, 550
Hepatic arterial perfusion scintigraphy, 205,
208–214
catheter insertion for, 208, 211f
clinical applications of, 213–214
image interpretation for, 214
methodology of, 213, 214b
Tc-99m MAA for, 209, 212
Hepatic parenchymal disease, 206f
Hepatocellular carcinoma
cholescintigraphy and, 189–190
tumor imaging for, 271
Hepatomegaly, 205b
Heterotopic bone formation, 141
Heterotopic gastric mucosa, 372–380
Barrett’s esophagus and, 380
diagnosis of
dosimetry in, 375
radiopharmaceuticals for, 374
uptake mechanisms in, 374–375
gastrointestinal duplications and, 379
Meckel’s diverticulum and, 376–379
retained gastric antrum and, 379–380
High-pass filter, 58
Histiocytosis, 128
INDEX 571
HIV. See Human immunodeficiency virus
HIV encephalopathy, 431
Hodgkin’s disease (HD)
Anne Arbor staging of, 270b
Ga-67 imaging of, 269–271
accuracy of, 269
residual masses and, 271
staging with, 269–270
NHL comparisons with, 269t
Rye classification of, 270t
Hot nose, 560
Human immunodeficiency virus (HIV), 431
Huntington’s chorea, 442
Hydrocephalus, 443–444
classifications of, 443t
image interpretation and, 444
symptoms of, 561
Hydrogen breath test, 382
Hydronephrosis
nonobstructed, 240f
obstructed, 240f
Hyperacute graft rejection, 243
clinical presentation of, 102–103
diagnosis of, 103
etiology of, 102b
imaging of, 103
pathology of, 102
skeletal scintigraphy of, 131–132
uptake distribution in, 141b
whole body view with, 143f
treatment of, 103
Hypokinesis, 495
I
I-123. See Iodine-123
I-131. See Iodine-131
I-131 OIH. See Iodine-131
orthoiodohippurate
Iatrogenic trauma, 135
Idiopathic interstitial pulmonary fibrosis,
429
Immunoscintigraphy, 286–294
clinical applications of
colorectal cancer and, 286–290
lung cancer and, 293–294
ovarian cancer and, 290
prostate cancer and, 290–293
methodology of
CEA-Scan protocol for, 290b
ProstaScint protocol for, 293b
Verluma and, 294
radiopharmaceuticals for, 286t
CEA-Scan and, 287–290
dosimetry of, 291t
OncoScint CR/OV and, 287–290
ProstaScint and, 291–292
Verluma and, 294
Immunosuppressed patients, 416–418
cytomegalovirus and, 417
diagnosis of, 417f
diffuse parenchymal uptake and, 417
focal pulmonary uptake and, 417
infection scintigraphy of, 417
intracerebral infection and, 417–418
Immunosuppressive drugs
complications with, 245–246
for renal transplant, 241
In-111. See Indium-111
In-111 oxine. See Indium-111 oxine
leukocytes
572 INDEX
In-111 pentreotide. See Indium-111
pentreotide
Indium-111 (In-111)
physical characteristics of, 5t, 12–13, 387t
production of, 4
thyroid studies with, 96
Indium-111 oxine leukocytes (In-111 oxine),
389–394
chemistry of, 390
dosimetry of, 393t
infection scintigraphy with
interpretation of, 394
methodology of, 392–394
infection seeking time of, 396b
quality control of, 391
radiolabeling of, 391, 391b
Indium-111 pentreotide (In-111 pentreotide)
chemistry of, 281, 281f
dosimetry of, 282, 282t
peptide receptor imaging with
accuracy of, 281–282, 281t
image interpretation with, 282
methodology of, 282
protocol of, 282b
pharmacokinetics of, 281
Infarct-avid scintigraphy, 503–506
clinical applications of, 505–506
false-positives in, 506b
indications for, 506b
interpretation of, 504–505
methodology of, 504, 504b
radiopharmaceuticals for, 503
Infection pathophysiology, 385–386, 386f
Infection scintigraphy, 384–418
clinical applications of, 413–431
abdominal and pelvic infections
and, 418
cardiovascular disease and, 413–414
diabetic foot and, 407–408
fever and, 418
immunosuppressed patients and,
416–418
infected joint prostheses and, 411–412
inflammatory bowel disease and, 413
intra-abdominal infection and, 412, 413t
malignant external otitis and, 418
osteomyelitis and, 404–407
pulmonary infections and, 414–416
renal disease and, 413
frequently asked questions regarding,
558–560
indications for
Ga-67 and, 405b
In-111 oxine and, 405b
Tc-99m HMPAO and, 405b
interpretation of
Ga-67, 388
In-111 oxine leukocyte and, 394
Tc-99m HMPAO and, 397
methodology of
Ga-67 and, 387–388, 392b
In-111 oxine leukocyte and, 392–394
Tc-99m HMPAO and, 396–397, 403b
radiopharmaceuticals for, 386–401
comparison of, 402t
FDG and, 399
Ga-67 and, 386–388
In-111 oxine leukocytes and, 389–394
investigational, 399–401
list of, 385b
localization mechanisms of, 387b
monoclonal antibodies and, 397
normal distribution of, 387t
Infection scintigraphy (Continued ) Iodine (Continued )
physical characteristics of, 387t characteristics of, 12
pitfalls with, 401b decay scheme for, 25f
Tc-99m HMPAO and, 394–397 half-value layers of, 36t
Inflammation physical characteristics of, 5t
abnormal processes of, 312f production of, 4
frequently asked questions regarding, quality control of, 12
558–560 thyroid cancer imaging with, 99b
pathophysiology of, 385–386, 386f thyroid scintigraphy with, 73–74
PET and, 307–310 for thyroid uptake studies, 76–79
terminology of, 385t Iodine-131 orthoiodohippurate (I-131 OIH)
Inflammatory bowel disease, 413 chemistry of, 217
Instruments dosimetry of, 227t
computational pharmacokinetics of, 217–218
data analysis with, 50 special considerations of, 218–219
data display with, 50–51 uptake mechanisms of, 218b, 218f
digital image creation with, 49–50 Iodine-131 tositumomab
frequently asked questions regarding, characteristics of, 296t
543–544 chemistry of, 296
gamma ray spectrometry and NHL therapy with
backscatter peak in, 38 indications for, 296
coincidence peaks in, 38 methodology of, 296–297, 297t
Compton scatter in, 38 results of, 297
Compton valley, edge, and plateau in, 38 toxicity of, 297
iodide escape peak in, 38 Ionization chambers, 35, 543
lead characteristic x-ray peak in, 38 Ischemia, physiology of, 458
photopeak in, 37–38 Ischemic nephropathy. See Vasomotor
imaging nephropathy
gamma scintillation cameras for, 39–49 Isobar, 21
rectilinear scanners for, 39 Isomer, 21
PET related, 63–66 Isomeric transition, 26–27
coincidence detection with, 64–66, 65f Isotone, 21
detector materials for, 64, 64t Isotope, 21
detector pairing and, 64
gantry size in, 64
ring detector for, 63f
K
spatial resolution of, 66
Kaposi’s sarcoma, 280, 559
radiation detection, 34–51
Kidneys
basic ionization chambers for, 35
anatomy of, 217f
Geiger-Müller counters for, 35
function of, 216–217
proportional counters for, 35
clearance rate and, 216
scintillation detection with, 35–37
measurement of, 252–255
for SPECT, 53
renal plasma flow and, 218f
Internal conversion
visualization of, 547
characteristic x-ray emission and, 27f
Kinin system, 385t
in radionuclide decay, 27
Krypton-81m, 13, 513
International System of Units (SI), 28t
Kveim-Siltzbach test, 416
Intestinal function, 380–383
protein-losing enteropathy and, 383
Schilling test of, 380–381 L
transit scintigraphy for
large bowel, 382–383 Lactation
small bowel, 381–382 radiopharmaceuticals and, 15, 15t
Intestinal transit scintigraphy, 381–383 thyroid scintigraphy during, 76
large bowel transit and, 382–383 Lag phase, 356–357
small bowel transit and, 382–383 Large bowel transit
Intra-abdominal infection, 412 radiography of, 382
Intracerebral infection, 417–418 scintigraphy of, 382–383
Inulin, 252 Lead characteristic x-ray peak, 38
Iodide escape peak, 38 Left bundle branch block
Iodine-123 (I-123) hypoperfusion and, 563
adult doses of, 545 myocardial perfusion scintigraphy and, 474
characteristics of, 12 Legg-Calvé-Perthes disease
physical characteristics of, 5t skeletal scintigraphy of, 143, 145
production of, 4 view of, 150f
quality control of, 12 Leukemia, 128
thyroid cancer imaging with, 99b Leukocytes
thyroid scintigraphy with, 74 accumulation of, 559
for thyroid uptake studies, 76–79 physiology of, 388–389
Iodine-123 metaiodobenzylguanidine, 506 polymorphonuclear, 385t
Iodine-126, 26f radiolabeled, 389–397
Iodine-131 (I-131) In-111 oxine and, 389–394
adult doses of, 545 Tc-99m fanolesomab and, 397
 INDEX 573

Leukocytes (Continued ) Lymphoma (Continued ) Monoclonal antibodies, 284–297

Tc-99m HMPAO and, 394–397 imaging with, 335–336 background of, 284–286
Tc-99m sulesomab and, 397 skeletal scintigraphy and, 128 IgG fragmentation and, 284f
Leukotrienes, 385t staging of, 336–337 immunoscintigraphy with
Line of stability, 23 Lymphoscintigraphy, 300–301 clinical applications of, 286–294
Lingual thyroid, 94 for breast cancer, 301 methodology of, 290b, 293b, 294
Liposomes, 401 frequently asked questions regarding, 556 radiopharmaceuticals for, 286t, 287–294,
Liver. See also Cavernous hemangiomas for melanoma, 300 291t, 397
anatomy of methodology of, 301 NHL therapy with, 294–297
normal, 201f radiopharmaceuticals for, 301 production of, 285f
SPECT image of, 203f–204f Lysosomal enzymes, 385t radiolabeling of, 286
vascular, 193f Monocytes
cholescintigraphy of, 187 inflammation and, 385t
focal defects in, 208b M leukocyte physiology and, 389
frequently asked questions regarding, Motilin, 357
548–551 Macrophages, 389 Movement disorders, 441–442
Liver hemangiomas, 550 Magnetic resonance MRCP. See Magnetic resonance
Liver pâté, 359 cholangiopancreatography (MRCP), 177 cholangiopancreatography
Liver-spleen scintigraphy, 195–200 Malignant external otitis, 418 Multigated acquisition. See Radionuclide
clinical applications of, 196 Mammography ventriculography
image interpretation of, 196–197 accuracy of, 277 Multinodular toxic goiter, 85–86
focal nodular hyperplasia and, 197, 200 scintigraphic, 278b Multiple endocrine neoplasias (MEN)
liver uptake and, 197, 208b Mass, 23 list of, 280
superior vena cava obstruction and, 197 Mass deficit, 23 neuroendocrine tumors and, 280–281
protocol for, 200b Mast cell, 385t Multiple myeloma, 127–128
Tc-99m sulfur colloid for, 195–196 Maximum intensity projection scan (MIPS), Musculoskeletal tumor, 345
dosimetry of, 196, 200t 61 Myocardial perfusion scintigraphy, 451–485
pharmacokinetics of, 195 Mean life, 30 coronary artery disease and, 457–485
preparation of, 195 Meckel’s diverticulum, 376–379 cardiac stress tests and, 458–459
Low-pass filter, 58 clinical manifestations of, 376 prognosis of, 482–485
Lung anatomy, 522f epidemiology of, 379b frequently asked questions regarding,
Lung cancer, 315–324 heterotopic gastric mucosa and, 376 561–564
diagnosis of location of, 558 interpretation of, 466–470
respiratory motion artifacts and, 318f scintigraphic diagnosis of attenuation and, 469–480
solitary pulmonary nodules in, 317, accuracy of, 377 coronary anatomy and, 466–467, 467f
319–320 false-positives in, 377, 382b extracardiac uptake and, 470
immunoscintigraphy for, 294 interpretation, 377 infarction and, 470, 474
lymph nodes and methodology of, 376–377 ischemia and, 470, 474
anatomy of, 324f–325f Medical event, 16, 541 left bundle branch block and, 474
classification of, 323b Melanoma multiple perfusion defects and, 474–477
transaxial diagram of, 326f lymphoscintigraphy for, 300–301 myocardial perfusion and, 466–467
osseous metastases and, 121 methodology of, 301 myocardial viability assessment and,
skeletal scintigraphy and, 125–126 radiopharmaceuticals for, 301 477–480
staging of PET for, 337–338 quality control and, 467–468
histologic classification for, 320b tumor imaging for, 271 rest v. stress and, 468–469
NSCLC and, 320–323 MEN. See Multiple endocrine neoplasias methodology of, 454–457
SCLC and, 323–324 Metabolic bone disease gated SPECT and, 457
TNM classification for, 322b diagnosis of, 547 physiology of ischemia and, 458
survival rates associated with, 293 skeletal scintigraphy for, 131–133, 139b planar imaging and, 454, 454f
treatment for, 294 hyperparathyroidism in, 131–132, 141b, SPECT and, 455–457
tumor imaging for, 271–272 143f Tc-99m protocol for, 456b
ventilation perfusion scintigraphy for, 535 osteoporosis in, 132, 144f Tl-201 protocol for, 456b
Lymph nodes Paget’s disease in, 132–133, 144f, 145f quantitative analysis of, 480–481
chest Metaiodobenzylguanidine (MIBG) radiopharmaceuticals for, 451–454
anatomy of, 324f–325f for adrenomedullary scintigraphy, 110 pharmacokinetics of, 452t
classification of, 323b thyroid studies with, 96 physical characteristics of, 451b
transaxial diagram of, 326f Methimazole, 546 single photon, 454
head and neck MIBG. See Metaiodobenzylguanidine Tc-99m MIBI and, 453
anatomy of, 330f MIPS. See Maximum intensity projection Technetium-99m tetrofosmin and,
transaxial diagram of, 331f scan 453–454
pelvic and abdominal Misadministration. See Medical event Tl-201 and, 451–453
anatomy of, 291f Mo-99. See Molybdenum-99 sensitivity/specificity of, 481–482
dissection of, 292t Moiré patterns, 48, 48f Myocarditis, 502
ProstaScint and, 292t Molybdenum-99 (Mo-99) Myositis ossificans, 149f
Lymphoceles, 251 decay curve of, 6f
Lymphocytes decay scheme of, 26
infection scintigraphy with, 401 generators of, 4, 6, 6t N
leukocyte physiology and, 389 quality control for, 7–9
Lymphoma. See also Hodgkin’s disease; Non- wet v. dry systems for, 7 N-13. See Nitrogen-13
Hodgkin’s lymphoma yield of, 6–7 Negatron decay, 25
Anne Arbor classification of, 336b legal limit of, 541 Negatrons
PET for, 333–337 physical characteristics of, 5t definition of, 25
detection with, 336 production of, 4 mass-energy interactions with, 30
574 INDEX
Neuroblastoma
adrenomedullary scintigraphy for, 111–112
skeletal scintigraphy and, 126
Neutrons
characteristics of, 20
mass-energy equivalence of, 23t
Neutrophils, 389
NHL. See Non-Hodgkin’s lymphoma
Nitrogen-13 (N-13), 5t
cardiac perfusion imaging with, 486
dosimetry of, 487t
Non-Hodgkin’s lymphoma (NHL)
Ga-67 imaging of, 269–271
accuracy of, 269
residual masses and, 271
staging with, 269–270
HD comparisons with, 269t
radiolabeled antibody therapy for, 294–297
indications for, 295–296
methodology of, 296–297, 296b, 297b
radiopharmaceuticals for, 295–297
results of, 296–297
view of, 271f
Non-small cell lung cancer (NSCLC)
staging of, 320–323
restaging in, 323
TNM classification for, 322b
therapy response assessment for, 323
Nonspecific polyclonal immunoglobulin, 400
Nonulcer dyspepsia, 358
NRC. See Nuclear Regulatory Commission
NSCLC. See Non-small cell lung cancer
Nuclear binding energy, 23
Nuclear pharmacy
dose calibrators and, 17–18
pyrogen testing and, 17
sterility and, 17
Nuclear reactors, 4
Nuclear Regulatory Commission (NRC)
medical event definition by, 16
radiation dose limits by, 16b
radionuclide purity standards of, 7–8
Nucleons, 20
Nuclides, 21
Nutcracker esophagus, 348
O surveillance images of, 144f
O-15. See Oxygen-15
Oncology. See Cancer
OncoScint CR/OV
accuracy of, 279–289
adverse effects of, 290
characteristics of, 288t
chemistry of, 287f
dosimetry of, 289, 291t
indications for, 288
pharmacokinetics of, 287
radiolabeled monoclonal antibody imaging
with, 287–290
Opsonization, 385t
Orbital electron, 22b
Ordered subset expectation maximization
(OSEM), 60
Organification, 545
OSEM. See Ordered subset expectation
maximization
Osseous metastases, 119–124
diagnosis of, 547
incidence of, 120b
scintigraphic patterns in, 121–124
cold lesions and, 124
Osseous metastases (Continued) Pancreatic cancer
extraskeletal uptake and, 124 images of, 335f
flare phenomenon and, 123 PET for, 332
multiple focal lesions and, 123b Papillary stenosis. See Sphincter of Oddi
solitary lesions and, 122–123 dysfunction
superscan and, 123 Parathyroid adenoma, 546
from specific tumors Parathyroid anatomy, 101–102
of breast carcinoma, 120–121, 124–125 Parathyroid scintigraphy, 101–105
of lung carcinoma, 121, 125–126 false-positives in, 546
of prostate carcinoma, 120 hyperparathyroidism and
workup algorithm for, 122f clinical presentation of, 102–103
Osteoarthritis, 547 diagnosis of, 103
Osteochondromas, 129 etiology of, 102b
Osteoid osteoma, 128–129 pathology of, 102
Osteomyelitis, 404–407 treatment of, 103
biopsy for, 405 parathyroid anatomy and, 101–102
infection scintigraphy for, 405–407 radiopharmaceuticals for, 103–105
accuracy of, 406t Particle accelerators, 4
bone scan in, 405–407 Pediatric patients, 15
Ga-67 in, 406 Pelvis
indications for, 405b infection scintigraphy of, 418
leukocyte scan in, 406–407 lymph nodes in
MRI for, 405, 406t anatomy of, 291f
pathology of, 404 dissection of, 292t
acute hematogenous osteomyelitis and, ProstaScint and, 292t
404–405 tumor imaging of, 272–273
children and, 406f Pentagastrin, 377
contiguous site infection and, 405 Peptide radiotherapy, 284
direct bone inoculation and, 405 Peptide receptor imaging, 280–284
radiography for, 405 In-111 pentreotide for
skeletal scintigraphy of, 147, 151–152, 154b accuracy of, 281–282, 281t
findings of, 151 chemistry of, 281, 281f
prosthesis evaluation with, 152 dosimetry of, 282, 282t
vascular supply and, 405f image interpretation with, 282
vertebral, 408–410 methodology of, 282
bone scan for, 408 pharmacokinetics of, 281
pathology of, 408 protocol of, 282b
scintigraphy for, 408, 410 neuroendocrine tumors and
Osteonecrosis categories of, 280, 280f
etiologies of, 149b MEN and, 280b
Legg-Calvé-Perthes disease and, 143, 145, somatostatin receptors and, 280–281
150f Tc-99m NeoTect for, 282, 284
sickle cell anemia and, 146–147, 151f dosimetry of, 284
skeletal scintigraphy and, 141–147 methodology of, 284
steroid-induced, 145–146 tumor therapy with, 284
Osteoporosis Perchlorate discharge test, 98–99
skeletal scintigraphy and, 132 Periodic table of elements, 5t
PET. See Positron emission tomography
Osteosarcoma, 279f Pharmacologic stress tests, 461–466
Ovarian cancer adenosine and, 461–465
histopathologic classification of, 342b comparison of, 465t
immunoscintigraphy for, 290 dipyridamole for, 461–465
OncoScint for, 290 dobutamine for, 465–466
PET for, 342–343, 343b, 344b Pheochromocytoma, 111
staging of, 342b Phosphorus-32 (P-32)
Oxygen-15 (O-15), 5t bone pain palliation with, 298
cardiac perfusion imaging with, 487 dosimetry of, 298t
dosimetry of, 487t physical characteristics of, 298t
Photoelectric absorption, 31–32, 31f
Photoelectron, 22b
P Photomultiplier tubes, 43–44, 44f
Photons
P-32. See Phosphorus-32 characteristics of, 22
Paget’s disease Compton scattering and, 542
monostotic, 144f patient-produced, 39–40, 40f
multifocal, 145f Photopeak, 37–38, 543
skeletal scintigraphy and, 132–133 Physics, 20–33
Pair production, 31 atoms and
Palliation of bone pain, 297–300 Bohr model of, 20–22, 21f
with P-32, 298 structure of, 20–21
with Re-186 HEDP, 299–300 constants in, 21b
with Sm-153, 298–299 electromagnetic radiation and, 22–23
with Sr-89, 298 energy spectrum of, 22f

Physics (Continued )
mathematics of, 22–23
frequently asked questions regarding,
541–543
mass-energy interactions in, 23
gamma rays and, 30–32
negatrons and, 30
positrons and, 30
x-rays and, 30–32
radionuclide radiation and, 23–28
alpha decay in, 24–25
auger electrons in, 28
beta minus decay in, 25
characteristic x-rays in, 28
electron capture in, 26
gamma ray emission in, 27–28
internal conversion in, 27
isomeric transition in, 26–27
line of stability in, 23
positron decay in, 25–26
Pick’s disease, 433f
PIOPED. See Prospective investigation of
pulmonary embolism diagnosis
Planck’s constant, 21b
Pleurodesis, 314f
Plummer’s disease. See Multinodular toxic
goiter
Poisson probability law, 32
Polymorphonuclear leukocytes, 385t
Positron decay, 25–26
Positron emission tomography (PET), 63–67,
302–345
cardiac system and, 485–490
coronary artery disease and, 487–488
myocardial viability and, 488–489
protocol for, 489–490, 489b
carrier molecules for, 304t
computed tomography with, 66–67
event detection possibilities with, 65f
frequently asked questions regarding,
544–545
historical perspectives on, 303
image reconstruction in, 66
instrumentation for, 63–66
coincidence detection with, 64–66, 65f
detector materials for, 64, 64t
detector pairing and, 64
gantry size in, 64
ring detector for, 63f
spatial resolution of, 66
interpretation of, 306–313
artifacts in, 310, 313
benign variants in, 307–313
cancer therapy and, 307–310
inflammation and, 307–310
malignant patterns in, 313–314
normal distribution and, 306–307
limitations of, 303
methodology of, 305–306, 305b
cancer protocol for, 305–306
patient scheduling concerns with, 308b
oncologic applications of, 304, 314–345,
553
bladder carcinoma and, 344–345
breast cancer and, 338–342
cervical carcinoma and, 343–344
colorectal carcinoma and, 331, 333b
esophageal carcinoma and, 326–330
gastrointestinal tumors and, 332–333
head and neck carcinoma and, 324–325,
328b, 328f
lung carcinoma and, 315–324
lymphomas and, 333–337
INDEX 575
Positron emission tomography (Continued ) Prostate cancer (Continued )
melanoma and, 337–338 lymph node anatomy and, 291f
musculoskeletal tumors and, 345 osseous metastases and, 120
ovarian cancer and, 342–343 PET for, 344
prostate cancer and, 344 staging of, 290
renal cell carcinoma and, 344–345 treatment of, 291
testicular carcinoma and, 344 Prostheses
thyroid carcinoma and, 325–326 infection in
patient motion and, 544 bone scan of, 411
radiopharmaceuticals for, 13–14, 14b, rate of, 411
303–305 scintigraphy of, 411–412
cardiac, 485–488, 486t skeletal scintigraphy and, 152
dosimetry of, 305, 305t Protein-losing enteropathy, 383
FDG and, 488 Proton bombardment, 4
intracellular kinetics of, 304f Protons
investigational agents in, 304–305 characteristics of, 20
N-13 and, 486 mass-energy equivalence of, 23t
O-15 water and, 487 rest mass of, 21b
physical characteristics of, 304t Psychiatric disorders, 442
Rb-82 and, 486–487 Pulmonary delayed washout, 564
Tc-99m MIBI and, 489–490 Pulmonary disease, RVG for, 502
SPECT and, 67 Pulmonary embolus, 508–509. See also
tumor uptake with, 320b Ventilation perfusion scintigraphy
Positrons frequently asked questions regarding,
annihilation of, 30f 564–566
definition of, 25 ventilation perfusion scintigraphy and,
kinetic energy of, 541 520–521
mass-energy interactions with, 30 Pulmonary infections
Postcholecystectomy pain syndrome drug reactions and, 416
causes of, 185b idiopathic interstitial pulmonary fibrosis
cholescintigraphy and, 182 and, 416
Pregnancy sarcoidosis and, 414–416
radiopharmaceuticals and, 15 scintigraphy of, 414
thyroid scintigraphy during, 76 Ga-67, 416
Proportional counters, 35 patterns of, 416
Propylthiouracil, 546 uptake in, 415b
Prospective investigation of pulmonary Pyrogens, 17
embolism diagnosis (PIOPED), 521–534 Pyrophosphate, 114f
criteria of, 523b
high-probability scans and, 522, 524f, 525f
intermediate probability scans and, 527, R
529f, 530f, 531f
low-probability scans and, 523, 525–527, Rad, 542
526f–527f, 528f Radiation accidents, 16–17, 17b
lung anatomy and, 522f Radiation detection, 34–51
mismatched defects and, 534, 535b computers and
normal scans and, 522 data analysis with, 50
recommendations by, 523b data display with, 50–51
special signs and, 527 digital image creation with, 48–50
fissure sign and, 532f gamma ray spectrometry and
stripe sign and, 532f backscatter peak in, 38
ventilation perfusion scintigraphy and coincidence peaks in, 38
accuracy of, 532, 534 Compton scatter in, 38
interpretation of, 532 Compton valley, edge, and plateau in, 38
Prostaglandins, 385t iodide escape peak in, 38
ProstaScint lead characteristic x-ray peak in, 38
accumulation of, 556 photopeak in, 37–38
adverse reactions to, 556 imaging and
chemistry of, 291 gamma scintillation cameras for, 39–49
dosimetry of, 291t rectilinear scanners for, 39
localization of, 555 instruments of, 34–51
normal distribution of, 292 basic ionization chambers for, 35
pelvic lymph node dissection and, 292t Geiger-Müller counters for, 35
pharmacokinetics of, 292 proportional counters for, 35
Prostate cancer, 290–293 scintillation detection and, 35–37
immunoscintigraphy for, 291–293 Radiation dosimetry, 18–19
accuracy of, 292 of common diagnostic procedures, 18t
dosimetry of, 292–293 formula for, 19
indications for, 292 Radiation safety
interpretation of, 293 accident procedures and, 16–17, 17b,
methodology of, 292, 293b 541
radiopharmaceuticals for, 291–292 dose limits and, 16b
incidence of, 290 procedures for, 14b
576 INDEX
Radioactive decay
definition of, 23
mathematics of
biological half-life in, 30
conversions in, 28b, 28t
decay constant in, 28–30
effective half-life in, 30
half-life in, 28–30
mean life in, 30
units in, 28
rate of, 542
statistics of, 32–33
transition energy in, 24
Radioactive inert gases, 13
Radioaerosols, 513
Radiochemicals, 1
Radioiodine therapy
dosage for, 100b
for goiter, 101
for Graves’ disease, 100–101
indications for, 100b
for thyroid cancer, 101
Radioiodines, 12–13, 545. See also Iodine-
123; Iodine-131
Radiolabeled leukocytes, 389–397
In-111 oxine and, 389–394
chemistry of, 390
dosimetry of, 393t
infection scintigraphy with, 392–394
infection seeking time of, 396b
quality control of, 391
radiolabeling of, 391, 394b
leukocyte physiology and, 388–389
Tc-99m fanolesomab and, 397
Tc-99m HMPAO and, 394–397
biological half-life of, 395
chemistry of, 394
dosimetry of, 393t
infection scintigraphy with,
396–397
infection-seeking time of, 396b
leukocyte labeling with, 394–395
Tc-99m sulesomab and, 397
Radiolabeled lymphocytes, 401
Radionuclide cystography, 256–258
dosimetry of, 258, 258t
interpretation of, 258
methodology of, 257–258, 258b
for vesicoureteral reflux, 256–257,
258f–259f
Radionuclide ventriculography (RVG),
490–503
acquisition techniques of, 492–495
equilibrium gated blood pool studies,
492–495, 493f
first-pass studies and, 492
protocol for, 495b
clinical applications of, 501–503
cardiac toxicity and, 502
cardiomyopathy and, 502
congenital heart disease and, 503
coronary artery disease and, 501
pulmonary disease and, 502
valvular heart disease and, 502
interpretation of, 495–501
diagnostic criteria for, 497b
ejection fraction and, 497–499
Fourier phase analysis and, 499–500
functional images and, 500–501
qualitative analysis and, 495
radiopharmaceuticals for
blood pool agents and, 491
first-pass agents and, 491–492
Radionuclides
carrier-free, 4
definition of, 3, 21
generators of, 4, 6, 540
quality control for, 7–9
wet v. dry systems for, 7
yield of, 6–7
half-lives of, 6t
mean life of, 30
positron-emitting, 5t
production of, 4
radiation of, 23–28
alpha decay in, 24–25
auger electrons in, 28
beta minus decay in, 25
characteristic x-rays in, 28
electron capture in, 26
gamma ray emission in, 27–28
internal conversion in, 27
isomeric transition in, 26–27
line of stability in, 23
positron decay in, 25–26
single-photon, 5t
tomography of, 52–53
Radiopharmaceuticals, 3–19
design characteristics of, 4
dispensing of, 14–16
adverse reactions and, 16
lactation and, 15, 15t
medical events during, 16
pediatric patients and, 15
pregnancy and, 15
safety procedures for, 14b
spills during, 16–17, 541
frequently asked questions regarding,
540–541
localization of, 3, 4t
preparation of, 9
purity of, 9t
radionuclides and
carrier-free, 4
characteristics of, 5t
definition of, 3
generators of, 4, 6–9
half-lives of, 6t
production of, 4
Radiotracers, 3
Radium-226, 24f
Ramp filter, 56, 56f
Rb-82. See Rubidium-82
RBC scintigraphy. See Red blood cell
scintigraphy
Re-186 HEDP. See Rhenium-186
hydroxyethylidene diphosphonate
REAL classification system. See Revised
European-American lymphoma
classification system
Rectilinear scanners, 39
Red blood cell (RBC) scintigraphy
cavernous hemangiomas and, 190–195
accuracy in, 193–195
diagnostic criteria for, 193
image interpretation of, 192–195
normal distribution in, 192–193
normal hepatic vascular anatomy of, 192
pathology of, 190–191
protocol for, 192, 192b
gastrointestinal bleeding and, 366–372
accuracy of, 371–372
cell labeling for, 366–368, 366b, 367b
diagnostic criteria of, 369–370
dosimetry of, 368, 368t
effectiveness of, 372
Red blood cell (RBC) scintigraphy (Continued)
interpretation of, 368–369
methodology of, 368, 368b
pitfalls of, 370–371, 376b
radionuclide ventriculography with,
491–492
Tc-99m for, 191
dosimetry of, 192, 192t
pharmacokinetics of, 191
physical characteristics of, 379t
Reflux scintigraphy, 351–354
accuracy of, 354
interpretation of, 352–354
methodology of, 352, 355b
quantification methods for, 355b
Rem, 542
Renal cell carcinoma, 344–345
Renal cortical imaging, 255–258
acute pyelonephritis and, 255
interpretation of, 255–256
methodology of, 255, 256b
radionuclide cystography for, 256–258
dosimetry of, 258, 258t
interpretation of, 258
methodology of, 257–258, 258b
vesicoureteral reflux and, 256–257
Renal disease, 413
Renal function measurement, 252–255
camera technique for, 254–255, 255f
ERPF for, 253
GFR for, 254
time interval for, 552
Renal osteodystrophy, 141f
Renal scintigraphy. See also Dynamic renal
scintigraphy
clinical indications for, 215, 216b
radiopharmaceuticals for, 216t, 217–223
dosimetry of, 223, 227t
I-131 OIH and, 217–219
Tc-99m DMSA for, 221–223
Tc-99m DTPA for, 219, 221
Tc-99m GH for, 221–223
Tc-99m MAG3 for, 221
uptake mechanisms of, 218t, 218f
Renal transplant, 239–252
allograft survival rates and, 241
immunosuppressive drug therapy for,
241
medical complications after, 244t
accelerated acute rejection and, 243–244
acute rejection and, 244, 247f–248f,
249f–250f, 251–252
chronic allograft nephropathy and,
244–245
hyperacute rejection and, 243
immunosuppressive drug toxicity and,
245–246
vasomotor nephropathy and, 242–243,
245f, 251–252
scintigraphic evaluation of
interpretation of, 251–252
methodology for, 251
sources for, 240
surgical complications after, 244t
arterial stenosis and, 246, 248
lymphoceles and, 251
ureteral obstruction and, 251
urinary leaks and, 248
vascular occlusion and, 246
surgical procedure of, 239, 244f
Renal vein thrombosis
dynamic renal scintigraphy for, 246
views of, 251f

Renovascular hypertension, 230–234
ACE inhibition renography and, 230–234
criteria for, 237b
image interpretation of, 232–233
imaging protocol for, 230, 232
indications for, 230
protocol for, 233b
renogram pattern with, 237f
reporting of, 233–234
causes of, 230
function curves with, 233f
pathophysiology of, 231f
Rest mass
of electron, 21b
of proton, 21b
Retained gastric antrum, 379–380
Revised European-American lymphoma
(REAL) classification system, 274b
Rhabdomyolysis
skeletal scintigraphy of, 141
view of, 149f
Rhenium-186 hydroxyethylidene
diphosphonate (Re-186 HEDP)
bone pain palliation with, 299–300
dosimetry of, 298t
Rim sign, 173–174, 174f, 548
Roentgen
Rubidium-82 (Rb-82), 5t, 13–14
cardiac perfusion imaging with, 486–487
dosimetry of, 487t
RVG. See Radionuclide ventriculography
Rye classification, 298t
S projection profile in, 56–57
Salivagram
aspiration study with, 356f
for gastroesophageal reflux, 354
Salivary secretions, 546
Samarium-153 (Sm-153)
bone pain palliation with, 298–299
dosimetry of, 542t
physical characteristics of, 298t
whole body scan with, 299f
Sarcoidosis, 414–416
coronary artery disease and, 485
diagnosis of, 415–416
bronchoalveolar lavage for, 416
Ga-67 scintigraphy for, 416
Kveim-Siltzbach test for, 416
radiography for, 415–416, 415b
presentation of, 415
scintigraphic patterns of, 559
symptoms of, 414–415
Schilling test, 380–381
Scintillation, 35–37
Scintimammography, 277–278
protocol for, 278b
views of, 279f
SCLC. See Small cell lung cancer
Scleroderma, 347–348
Scrotal scintigraphy, 258–260
interpretation of, 260
methodology of, 260
radiopharmaceuticals for, 260
for testicular torsion, 259–260
Secular equilibrium generators, 540
Septic joint, 154b
Serum creatinine, 252
Shin splints
diagnosis of, 548
skeletal scintigraphy of, 139
view of, 149f
Shrunken lung sign, 565
SI. See International System of Units
Sickle cell anemia
skeletal scintigraphy of, 146–147
view of, 151f
Sincalide
cholescintigraphy with, 164, 165b
clinical indications for, 164b
infusion methodology for, 549
SOD and, 549
Single-photon absorptiometry (SPA), 153–154
Single-photon emission computed
tomography (SPECT), 53–63
attenuation correction for, 58, 544
frequently asked questions regarding,
544–545
image acquisition for, 53–55
angular sampling interval for, 54
arc of, 54–55
collimator selection for, 53–54
issues in, 54b
matrix size in, 54
orbit and, 54
OSEM for, 60
patient factors in, 554
time of, 55
image reconstruction with, 55–60
angular projection in, 56
backprojection in, 57
filters for, 57–59, 544
Fourier transformation in, 56
frequency domain in, 56, 59, 59f
iterative methods for, 59–60
Nyquist frequency in, 57
ray sum in, 57
spatial domain in, 56
image reformatting in, 60–61
instrumentation for, 53
MIPS for, 61
patient motion and, 544
PET and, 67
quality assurance for, 61–63, 63b
Skeletal scintigraphy, 113–158
of bone marrow, 152–153
bone mineral measurement with, 152–158
applications of, 155
classification of, 157
DPA for, 154
DXA for, 154
SPA for, 153–154
clinical applications of
breast cancer tumors and, 124–125
epithelial tumors and, 126
lung cancer tumors and, 125–126
metabolic bone disease and, 131–133,
139b
metastatic disease and, 119–124
neuroblastoma and tumors and, 126
primary tumors and, 127–129, 135t
trauma and, 133–141
imaging protocol of, 115–117
three-phase, 116b
whole body survey, 116b
neonates and, 548
normal radiotracer distribution in, 117–119
osteomyelitis and, 147, 151–152
findings of, 151
prosthesis evaluation and, 152
osteonecrosis and, 141–147
etiologies of, 149b
Legg-Calvé-Perthes disease and, 143, 145
sickle cell anemia and, 146–147
steroid-induced, 145–146
INDEX 577
Skeletal scintigraphy (Continued)
radiopharmaceuticals for
dosimetry of, 115, 116t
history of, 114
pharmacokinetics of, 115
preparation of, 114–115
retention of, 546
three-phase, 154b
Skeletal trauma, 133–141
activity-induced enthesopathy and, 139
from child abuse, 135, 137
complex regional pain syndrome and,
137–138
fracture detection and, 133–135
heterotopic bone formation and, 141
iatrogenic trauma and, 135
rhabdomyolysis and, 141, 149f
shin splints and, 139, 149f
stress fractures and, 138–139
Sm-153. See Samarium-153
Small bowel barium follow-through study,
382
Small bowel transit, 381–382
barium follow-through study for, 382
diseases associated with, 381–382
hydrogen breath test for, 382
scintigraphy for, 382
Small cell lung cancer (SCLC), 323–324
SOD. See Sphincter of Oddi dysfunction
Sodium iodide crystals
detection efficiency of, 543
thallium impurity in, 543
Soft tissue sarcomas
PET for, 345
tumor imaging for, 273
Somatostatin
chemistry of, 280, 281f
neuroendocrine tumors and, 280–281
SPA. See Single-photon absorptiometry
SPECT. See Single-photon emission computed
tomography
Speed of light, 21b
Sphincter of Oddi dysfunction (SOD)
cholescintigraphy protocol for, 186b
diagnostic patterns of, 550
postoperative, 184–185
sequential images of, 187f
sincalide and, 549
Spleen scintigraphy, 200–201, 205
Spondylosis, 148f
Sr-89. See Strontium-89
Standard deviation, 543
Stannous ions, 541
Star artifact, 541
Sterility, 17
Steroids, 145–146
Stomach, 355–357
anatomy of, 356f, 557–558
antrum in, 355–356
fundus in, 355
gastric emptying of, 357b, 357f
lag phase of, 356–357
motilin in, 357
Stress fractures, 138–139
Stress perfusion scintigraphy, 459
Stripe sign, 527, 532f, 565
Stroke
brain death scintigraphy and, 435–438
images from, 437f–438f
interpretation of, 436, 438
methodology of, 436, 439b
radiopharmaceuticals for, 435–436
PET for, 431–435
SPECT for, 431–435
578 INDEX
Strontium-89 (Sr-89)
bone pain palliation with, 298
dosimetry of, 298t
physical characteristics of, 298t
toxicity of, 298
Struma ovarii, 87
Substernal goiter, 93
Superior vena cava syndrome
liver-spleen scintigraphy for, 197
view of, 208f
Surgical shunt patency, 444–445
hydrocephalus and, 444
methodology of, 445b
Swinging heart sign, 527
Synthroid, 546
T chemistry of, 394
T3 suppression test, 97
Tardokinesis, 495
Tc-99m. See Technetium-99m
Tc-99m DMSA. See Technetium-99m
dimercaptosuccinic acid
Tc-99m DTPA. See Technetium-99m
diethylenetriamine pentaacetic acid
Tc-99m GH. See Technetium-99m
glucoheptonate
Tc-99m HMPAO. See Technetium-99m
hexamethyl-propylene-amine oxime
Tc-99m MAA. See Technetium-99m
macroaggregated albumin
Tc-99m MAG3. See Technetium-99m
mercaptoacetythiglycine
Tc-99m MIBI. See Technetium-99m
sestamibi
Tc-99m NeoTect. See Technetium-99m
depreotide
Tc-99m SC. See Technetium-99m sulfur
colloid
Technetium-99m (Tc-99m)
agents with, 10t
applications of, 10t
buildup of, 541
chemistry and, 9, 161f
for cholescintigraphy, 160–161
generators of, 6, 6t
parathyroid scintigraphy with, 103–105
pharmacokinetics of, 162t
physical characteristics of, 5t, 387t
decay curve and, 8t, 29f
half-value layers and, 36t
ingrowth curve and, 6f
preparation of, 10f–11f
quality assurance with, 9–12
for RBC scintigraphy, 191–195
for skeletal scintigraphy, 114–115
thyroid studies with, 95
scintigraphic, 74–75
uptake, 79–80
Technetium-99m ciprofloxacin
infection scintigraphy with, 400–401
vertebral osteomyelitis and, 410
Technetium-99m depreotide (Tc-99m
NeoTect)
development of, 282
dosimetry of, 284
for peptide receptor imaging, 282, 284
Technetium-99m diethylenetriamine
pentaacetic acid (Tc-99m DTPA)
chemistry of, 219
dosimetry of, 227t
pharmacokinetics of, 221
Technetium-99m diethylenetriamine Thallium-201 (Tl-201)
(Continued) chemistry of, 274
ventilation perfusion scintigraphy with, dosimetry of, 277
512, 512t, 514–515, 514b electron capture in, 26f
Technetium-99m dimercaptosuccinic acid half-value layers of, 36t
(Tc-99m DMSA) myocardial perfusion scintigraphy with,
dosimetry of, 227t 451–453
pharmacokinetics of, 221–222 pharmacokinetics of, 274, 452t
radiolabeling of, 221 physical characteristics of, 5t, 13, 451b
Technetium-99m fanolesomab, 397 production of, 4
Technetium-99m glucoheptonate (Tc-99m GH) thyroid studies with, 95
dosimetry of, 227t tumor uptake mechanisms of, 276b
pharmacokinetics of, 221–222 Thorax, 547
radiolabeling of, 221 Thrombosis
Technetium-99m hexamethyl-propylene- arterial, 246
amine oxime (Tc-99m HMPAO) deep vein, 536–539
biological half-life of, 395 renal vein, 246, 251f
Thyroid
dosimetry of, 393t activity patterns of, 311f
infection scintigraphy with anatomy of, 72, 72f
interpretation of, 397 frequently asked questions regarding,
methodology of, 396–397 545–546
infection-seeking time of, 396b hyperfunction of, 85b
leukocyte labeling with, 394–395 iodine metabolism in, 72, 73f, 74f
Technetium-99m macroaggregated albumin organification and, 72
(Tc-99m MAA) thyroid-pituitary feedback mechanism and,
hepatic arterial perfusion scintigraphy 72–73, 74f
with, 209, 212 tissue origin of, 545
ventilation perfusion scintigraphy with, TSH and, 72
511, 515, 515b Thyroid cancer
Technetium-99m mercaptoacetythiglycine imaging of, 553–554
(Tc-99m MAG3) nodules and, 90b
dosimetry of, 227t PET for, 325–326
pharmacokinetics of, 221 radioiodine therapy for, 101
radiolabeling of, 221 thyroid nodules and, 90b
Technetium-99m sestamibi (Tc-99m MIBI) thyroid scintigraphy and, 82–83, 94–95
chemistry of, 274, 276 tumor imaging for, 280
dosimetry of, 277 Thyroid nodules, 88–92
myocardial perfusion scintigraphy with, cancer risk with, 90b
453 cold, 90–91
pharmacokinetics of, 276, 452t differential diagnosis of, 90b
physical characteristics of, 451b discordant, 92, 92f
tumor uptake mechanisms of, 276–277, fine needle aspiration for, 89
276b hot, 91
Technetium-99m sulesomab, 397 indeterminate, 91
Technetium-99m sulfur colloid thyroid scintigraphy for, 89–90
(Tc-99m SC) ultrasonography for, 89
gastrointestinal bleeding scintigraphy with, Thyroid scintigraphy, 80–83
364–366 cancer scans with, 82–83
dosimetry of, 368, 368t clinical indications for, 87–95
effectiveness of, 372 cancer and, 94–95
interpretation of, 365–366 ectopic thyroid tissue and, 73–74
methodology for, 365, 365b goiter and, 92–93
time-activity curves of, 365f thyroid nodules and, 88–92
views of, 365f, 366f thyroiditis and, 94
for liver-spleen imaging, 195–200 I-123 scans with, 80–81, 81b
physical characteristics of, 379t I-131 scans with, 81–82
Technetium-99m tetrofosmin normal v. abnormal, 88
chemistry of, 277 Tc-99 scans with, 80–81, 81b
dosimetry of, 277 Thyroid stimulating hormone (TSH), 72
myocardial perfusion scintigraphy with, Thyroid studies, 71–101. See also Thyroid
453–454 scintigraphy;Thyroid uptake studies
pharmacokinetics of, 277, 452t classical, 96–99
physical characteristics of, 451b perchlorate discharge test and, 98–99
tumor uptake mechanisms of, 277 T3 suppression test and, 97
Testicular carcinoma, 344 TSH stimulation test and, 97–98
Testicular torsion, 258–262 radiopharmaceuticals for, 73–76
in children, 259 dosimetry of, 76t
in neonates, 259 FDG and, 95–96
scrotal scintigraphy for, 260 I-123 and, 74
testicular viability after, 260 I-131 and, 73–74
ultrasound for, 258 In-111, 96
views of, 260f, 261f MIBG and, 96

Thyroid studies (Continued )
physical characteristics of, 76t
precautions with, 76
selection of, 75–76
Tc-99m and, 74–75, 95
Tl-201 and, 95
Tc-99m uptake for, 79–80
thyroid anatomy and, 72, 72f
thyroid physiology and, 72–73, 73f
Thyroid uptake studies, 76–79, 83–87
clinical indications for, 83–87
amiodarone-induced thyrotoxicosis
and, 87
Graves’ disease and, 85
Hashitoxicosis and, 86
iodine-induced thyrotoxicosis and,
86–87
rare causes of, 86
single autonomously functioning thyroid
nodule and, 86
struma ovarii and, 87
subacute thyroiditis and, 86
thyrotoxicosis and, 83–85
thyrotoxicosis factitia and, 87
differential diagnosis with, 546
factors affecting, 77b
indications for, 77, 78b
methodology of, 77–79
normal values in, 79
probes for, 79f
Thyroiditis
chronic, 86
subacute, 86, 87f
thyroid scintigraphy and, 94
thyroid uptake studies and, 86
Thyrotoxicosis
clinical diagnosis of, 83
clinical frequency of, 78b
differential diagnosis of, 78b, 83
iodine-induced, 86–87
radioiodine uptake and, 84–85
thyroid hyperfunction and, 85b
thyroid uptake studies of, 77, 87
Tl-201. See Thallium-201
T-lymphocytes, 385t
TNM. See Tumor, node, metastasis
classification system
Transient equilibrium, 540
Transition energy, 24
Trauma
head, 442
iatrogenic, 135
skeletal, 133–141
activity-induced enthesopathy and, 139
from child abuse, 135, 137
complex regional pain syndrome and,
137–138
fracture detection and, 133–135
heterotopic bone formation and, 141
iatrogenic trauma and, 135
rhabdomyolysis and, 141, 149f
shin splints and, 139, 149f
stress fractures and, 138–139
TSH. See Thyroid stimulating hormone
TSH stimulation test, 97–98
Tumor imaging, 264–280
Cerebral perfusion imaging for, 439, 441
clinical applications of
abdominal and pelvic cancers and,
272–273
bone and soft tissue tumors and, 278
breast cancer and, 277–278
HD and, 269–271
Tumor imaging (Continued )
head and neck cancers and, 272
hepatocellular carcinoma, 271
Kaposi’s sarcoma and, 280
lung cancer and, 271–272
malignant melanoma and, 271
NHL and, 269–271
soft tissue sarcomas and, 273
thyroid cancer and, 280
interpretation of, 267–268
methodology of, 266–267, 267b, 277
PET limitations and, 554
radiopharmaceuticals for, 264–266,
273–277
dosimetry of, 267t, 277
Ga-67 and, 264–266, 265f, 266b
overview of, 264b
physical characteristics of, 265t
Tc-99m MIBI and, 274, 276–277, 276b
Tc-99m tetrofosmin and, 277
Tl-210 and, 274, 275b
sensitivity of, 269t
Tumor, node, metastasis (TNM) classification
system
for colorectal carcinoma, 333b
for lung cancer, 322b
Tuttle acid reflux test, 351
U 529f, 530f, 531f
Ulcerative colitis, 413
Ulcers, 362
Ultrasonography
for acute cholecystitis, 169–170, 170b
for biliary duct obstruction, 177
for breast cancer, 277
for cavernous hemangiomas, 191
for testicular torsion, 259
for thyroid nodules, 89
for urinary tract obstruction, 234
Urea breath test, 363
Ureteral obstruction, 251
Urinary leaks, 248
Urinary tract obstruction, 234–239
diuretic renography for, 234, 237–239
diuretic administration for, 238–239
interpretation of, 239
limitations of, 239
patient preparation for, 237–238
protocol for, 238b
radiopharmaceuticals for, 237
time-activity curves of, 241f
hydronephrosis and
nonobstructed, 240f
obstructed, 240f
ultrasonography for, 234
Whitaker test for, 234
Uveoparotid fever, 416
V grading of, 259f
Valence electron, 22b, 542
Valvular heart disease, 502
Vascular occlusion, 246
Vasoamines, 385t
Vasomotor nephropathy
dynamic renal scintigraphy for, 242–243,
251–252
time-activity curve of, 245f
Ventilation perfusion scintigraphy, 509–521
accuracy of, 532, 534
INDEX 579
Ventilation perfusion scintigraphy (Continued)
clinical applications of
adult respiratory distress syndrome and,
535–536
quantitative lung scans and, 535
frequently asked questions regarding,
564–566
interpretation of, 515–521
analysis and, 535t
differential diagnosis and, 534
lung perfusion abnormalities and, 520b
nonsegmental defects and, 523b
perfusion defects and, 519b
perfusion scintigrams and, 515, 517
PIOPED approach to, 532
pulmonary embolisms and, 520–521
sensitivity/specificity and, 535t
ventilation abnormalities and, 518–520
ventilation scintigrams and, 517–518
methodology of, 513–515
Tc-99m DPTA and, 514–515, 514b
Tc-99m MAA and, 515, 515b
Xe-133 ventilation and, 514, 514b
physiological parameters and, 509
PIOPED and, 521–534
criteria of, 523b
high-probability scans and, 522, 524f,
525f
intermediate probability scans and, 527,
low-probability scans and, 523, 525–527,
526f–527f, 528f
lung anatomy and, 522f
mismatched defects and, 534, 535b
normal scans and, 522
recommendations by, 523b
special signs and, 527, 532f
radiopharmaceuticals for, 510–513
dosimetry of, 513t
historical, 511b
perfusion, 510–511
physical characteristics of, 512t
radioactive gases and, 512–513
radioaerosols and, 513
ventilation, 512
terminology of, 521b
Ventriculography. See Radionuclide
ventriculography
Verluma
chemistry of, 294
dosimetry of, 291t
pharmacokinetics of, 294
Vertebral osteomyelitis, 408–410
bone scan for, 408
pathology of, 408
scintigraphy for
FDG, 410
Ga-67, 410
leukocyte, 408, 410
Technetium-99m ciprofloxacin, 410
Vesicoureteral reflux
radionuclide cystography for, 256–257
view of, 258f–259f
Vocal cord paralysis, 309f
W
Webster’s rule, 541, 551
Whitaker test
for urinary tract obstruction, 234
view of, 238f
580 INDEX

X X-rays (Continued)
definition of, 23, 542
Yttrium-90 ibritumomab tiuxetan (Continued)
NHL therapy with
Xe-133. See Xenon-133 pair production with, 31 indications for, 295
Xenon-127, 5t, 13, 513 photoelectric absorption with, 31–32, 31f methodology of, 296, 296t
Xenon-133 (Xe-133) results of, 296
physical characteristics of, 5t, 13, 512t toxicity of, 296
production of, 4 Y
ventilation perfusion scintigraphy with,
512–513, 514, 514b Yttrium-90 ibritumomab tiuxetan
X-rays characteristics of, 295
Compton effect with, 32 chemistry of, 295
Compton scattering with, 32, 32f dosimetry of, 296, 296t